-
Ulcerative Colitis Mayo Endoscopic Scoring Classification with Active Learning and Generative Data Augmentation
Authors:
Ümit Mert Çağlar,
Alperen İnci,
Oğuz Hanoğlu,
Görkem Polat,
Alptekin Temizel
Abstract:
Endoscopic imaging is commonly used to diagnose Ulcerative Colitis (UC) and classify its severity. It has been shown that deep learning based methods are effective in automated analysis of these images and can potentially be used to aid medical doctors. Unleashing the full potential of these methods depends on the availability of large amount of labeled images; however, obtaining and labeling thes…
▽ More
Endoscopic imaging is commonly used to diagnose Ulcerative Colitis (UC) and classify its severity. It has been shown that deep learning based methods are effective in automated analysis of these images and can potentially be used to aid medical doctors. Unleashing the full potential of these methods depends on the availability of large amount of labeled images; however, obtaining and labeling these images are quite challenging. In this paper, we propose a active learning based generative augmentation method. The method involves generating a large number of synthetic samples by training using a small dataset consisting of real endoscopic images. The resulting data pool is narrowed down by using active learning methods to select the most informative samples, which are then used to train a classifier. We demonstrate the effectiveness of our method through experiments on a publicly available endoscopic image dataset. The results show that using synthesized samples in conjunction with active learning leads to improved classification performance compared to using only the original labeled examples and the baseline classification performance of 68.1% increases to 74.5% in terms of Quadratic Weighted Kappa (QWK) Score. Another observation is that, attaining equivalent performance using only real data necessitated three times higher number of images.
△ Less
Submitted 10 November, 2023;
originally announced November 2023.
-
Biomedical image analysis competitions: The state of current participation practice
Authors:
Matthias Eisenmann,
Annika Reinke,
Vivienn Weru,
Minu Dietlinde Tizabi,
Fabian Isensee,
Tim J. Adler,
Patrick Godau,
Veronika Cheplygina,
Michal Kozubek,
Sharib Ali,
Anubha Gupta,
Jan Kybic,
Alison Noble,
Carlos Ortiz de Solórzano,
Samiksha Pachade,
Caroline Petitjean,
Daniel Sage,
Donglai Wei,
Elizabeth Wilden,
Deepak Alapatt,
Vincent Andrearczyk,
Ujjwal Baid,
Spyridon Bakas,
Niranjan Balu,
Sophia Bano
, et al. (331 additional authors not shown)
Abstract:
The number of international benchmarking competitions is steadily increasing in various fields of machine learning (ML) research and practice. So far, however, little is known about the common practice as well as bottlenecks faced by the community in tackling the research questions posed. To shed light on the status quo of algorithm development in the specific field of biomedical imaging analysis,…
▽ More
The number of international benchmarking competitions is steadily increasing in various fields of machine learning (ML) research and practice. So far, however, little is known about the common practice as well as bottlenecks faced by the community in tackling the research questions posed. To shed light on the status quo of algorithm development in the specific field of biomedical imaging analysis, we designed an international survey that was issued to all participants of challenges conducted in conjunction with the IEEE ISBI 2021 and MICCAI 2021 conferences (80 competitions in total). The survey covered participants' expertise and working environments, their chosen strategies, as well as algorithm characteristics. A median of 72% challenge participants took part in the survey. According to our results, knowledge exchange was the primary incentive (70%) for participation, while the reception of prize money played only a minor role (16%). While a median of 80 working hours was spent on method development, a large portion of participants stated that they did not have enough time for method development (32%). 25% perceived the infrastructure to be a bottleneck. Overall, 94% of all solutions were deep learning-based. Of these, 84% were based on standard architectures. 43% of the respondents reported that the data samples (e.g., images) were too large to be processed at once. This was most commonly addressed by patch-based training (69%), downsampling (37%), and solving 3D analysis tasks as a series of 2D tasks. K-fold cross-validation on the training set was performed by only 37% of the participants and only 50% of the participants performed ensembling based on multiple identical models (61%) or heterogeneous models (39%). 48% of the respondents applied postprocessing steps.
△ Less
Submitted 12 September, 2023; v1 submitted 16 December, 2022;
originally announced December 2022.
-
Pair binding and enhancement of pairing correlations in asymmetric Hubbard ladders
Authors:
Anas Abdelwahab,
Gökmen Polat,
Eric Jeckelmann
Abstract:
Asymmetric two-leg Hubbard ladders with different on-site interactions $U_y$ and hop**s $t_y$ on each leg are investigated using the density matrix renormalization group method and exact diagonalizations. The pairing found in symmetric ladders is robust against the introduction of the leg asymmetry. When studying pairing, one-band Hubbard ladder models are better described as one-dimensional cor…
▽ More
Asymmetric two-leg Hubbard ladders with different on-site interactions $U_y$ and hop**s $t_y$ on each leg are investigated using the density matrix renormalization group method and exact diagonalizations. The pairing found in symmetric ladders is robust against the introduction of the leg asymmetry. When studying pairing, one-band Hubbard ladder models are better described as one-dimensional correlated two-band models than as sublattices of higher dimensional systems. The asymmetric Hubbard ladder provides us with a simple model for studying pairing in the crossover regime between charge transfer and Mott insulators.
△ Less
Submitted 8 December, 2022; v1 submitted 12 September, 2022;
originally announced September 2022.
-
Assessing generalisability of deep learning-based polyp detection and segmentation methods through a computer vision challenge
Authors:
Sharib Ali,
Noha Ghatwary,
Debesh Jha,
Ece Isik-Polat,
Gorkem Polat,
Chen Yang,
Wuyang Li,
Adrian Galdran,
Miguel-Ángel González Ballester,
Vajira Thambawita,
Steven Hicks,
Sahadev Poudel,
Sang-Woong Lee,
Ziyi **,
Tianyuan Gan,
ChengHui Yu,
JiangPeng Yan,
Doyeob Yeo,
Hyunseok Lee,
Nikhil Kumar Tomar,
Mahmood Haithmi,
Amr Ahmed,
Michael A. Riegler,
Christian Daul,
Pål Halvorsen
, et al. (7 additional authors not shown)
Abstract:
Polyps are well-known cancer precursors identified by colonoscopy. However, variability in their size, location, and surface largely affect identification, localisation, and characterisation. Moreover, colonoscopic surveillance and removal of polyps (referred to as polypectomy ) are highly operator-dependent procedures. There exist a high missed detection rate and incomplete removal of colonic pol…
▽ More
Polyps are well-known cancer precursors identified by colonoscopy. However, variability in their size, location, and surface largely affect identification, localisation, and characterisation. Moreover, colonoscopic surveillance and removal of polyps (referred to as polypectomy ) are highly operator-dependent procedures. There exist a high missed detection rate and incomplete removal of colonic polyps due to their variable nature, the difficulties to delineate the abnormality, the high recurrence rates, and the anatomical topography of the colon. There have been several developments in realising automated methods for both detection and segmentation of these polyps using machine learning. However, the major drawback in most of these methods is their ability to generalise to out-of-sample unseen datasets that come from different centres, modalities and acquisition systems. To test this hypothesis rigorously we curated a multi-centre and multi-population dataset acquired from multiple colonoscopy systems and challenged teams comprising machine learning experts to develop robust automated detection and segmentation methods as part of our crowd-sourcing Endoscopic computer vision challenge (EndoCV) 2021. In this paper, we analyse the detection results of the four top (among seven) teams and the segmentation results of the five top teams (among 16). Our analyses demonstrate that the top-ranking teams concentrated on accuracy (i.e., accuracy > 80% on overall Dice score on different validation sets) over real-time performance required for clinical applicability. We further dissect the methods and provide an experiment-based hypothesis that reveals the need for improved generalisability to tackle diversity present in multi-centre datasets.
△ Less
Submitted 24 February, 2022;
originally announced February 2022.
-
Evaluation and Analysis of Different Aggregation and Hyperparameter Selection Methods for Federated Brain Tumor Segmentation
Authors:
Ece Isik-Polat,
Gorkem Polat,
Altan Kocyigit,
Alptekin Temizel
Abstract:
Availability of large, diverse, and multi-national datasets is crucial for the development of effective and clinically applicable AI systems in the medical imaging domain. However, forming a global model by bringing these datasets together at a central location, comes along with various data privacy and ownership problems. To alleviate these problems, several recent studies focus on the federated…
▽ More
Availability of large, diverse, and multi-national datasets is crucial for the development of effective and clinically applicable AI systems in the medical imaging domain. However, forming a global model by bringing these datasets together at a central location, comes along with various data privacy and ownership problems. To alleviate these problems, several recent studies focus on the federated learning paradigm, a distributed learning approach for decentralized data. Federated learning leverages all the available data without any need for sharing collaborators' data with each other or collecting them on a central server. Studies show that federated learning can provide competitive performance with conventional central training, while having a good generalization capability. In this work, we have investigated several federated learning approaches on the brain tumor segmentation problem. We explore different strategies for faster convergence and better performance which can also work on strong Non-IID cases.
△ Less
Submitted 12 April, 2022; v1 submitted 16 February, 2022;
originally announced February 2022.
-
Class Distance Weighted Cross-Entropy Loss for Ulcerative Colitis Severity Estimation
Authors:
Gorkem Polat,
Ilkay Ergenc,
Haluk Tarik Kani,
Yesim Ozen Alahdab,
Ozlen Atug,
Alptekin Temizel
Abstract:
In scoring systems used to measure the endoscopic activity of ulcerative colitis, such as Mayo endoscopic score or Ulcerative Colitis Endoscopic Index Severity, levels increase with severity of the disease activity. Such relative ranking among the scores makes it an ordinal regression problem. On the other hand, most studies use categorical cross-entropy loss function to train deep learning models…
▽ More
In scoring systems used to measure the endoscopic activity of ulcerative colitis, such as Mayo endoscopic score or Ulcerative Colitis Endoscopic Index Severity, levels increase with severity of the disease activity. Such relative ranking among the scores makes it an ordinal regression problem. On the other hand, most studies use categorical cross-entropy loss function to train deep learning models, which is not optimal for the ordinal regression problem. In this study, we propose a novel loss function, class distance weighted cross-entropy (CDW-CE), that respects the order of the classes and takes the distance of the classes into account in calculation of the cost. Experimental evaluations show that models trained with CDW-CE outperform the models trained with conventional categorical cross-entropy and other commonly used loss functions which are designed for the ordinal regression problems. In addition, the class activation maps of models trained with CDW-CE loss are more class-discriminative and they are found to be more reasonable by the domain experts.
△ Less
Submitted 12 June, 2022; v1 submitted 9 February, 2022;
originally announced February 2022.
-
ARFED: Attack-Resistant Federated averaging based on outlier elimination
Authors:
Ece Isik-Polat,
Gorkem Polat,
Altan Kocyigit
Abstract:
In federated learning, each participant trains its local model with its own data and a global model is formed at a trusted server by aggregating model updates coming from these participants. Since the server has no effect and visibility on the training procedure of the participants to ensure privacy, the global model becomes vulnerable to attacks such as data poisoning and model poisoning. Althoug…
▽ More
In federated learning, each participant trains its local model with its own data and a global model is formed at a trusted server by aggregating model updates coming from these participants. Since the server has no effect and visibility on the training procedure of the participants to ensure privacy, the global model becomes vulnerable to attacks such as data poisoning and model poisoning. Although many defense algorithms have recently been proposed to address these attacks, they often make strong assumptions that do not agree with the nature of federated learning, such as assuming Non-IID datasets. Moreover, they mostly lack comprehensive experimental analyses. In this work, we propose a defense algorithm called ARFED that does not make any assumptions about data distribution, update similarity of participants, or the ratio of the malicious participants. ARFED mainly considers the outlier status of participant updates for each layer of the model architecture based on the distance to the global model. Hence, the participants that do not have any outlier layer are involved in model aggregation. We have performed extensive experiments on diverse scenarios and shown that the proposed approach provides a robust defense against different attacks. To test the defense capability of the ARFED in different conditions, we considered label flip**, Byzantine, and partial knowledge attacks for both IID and Non-IID settings in our experimental evaluations. Moreover, we proposed a new attack, called organized partial knowledge attack, where malicious participants use their training statistics collaboratively to define a common poisoned model. We have shown that organized partial knowledge attacks are more effective than independent attacks.
△ Less
Submitted 16 June, 2023; v1 submitted 8 November, 2021;
originally announced November 2021.
-
Exclusive $B \to X_u \ell ν_\ell$ Decays with Hadronic Full-event-interpretation Tagging in 62.8 fb$^{-1}$ of Belle II Data
Authors:
Belle II Collaboration,
F. Abudinén,
I. Adachi,
R. Adak,
K. Adamczyk,
L. Aggarwal,
P. Ahlburg,
H. Ahmed,
J. K. Ahn,
H. Aihara,
N. Akopov,
A. Aloisio,
F. Ameli,
L. Andricek,
N. Anh Ky,
D. M. Asner,
H. Atmacan,
V. Aulchenko,
T. Aushev,
V. Aushev,
T. Aziz,
V. Babu,
S. Bacher,
H. Bae,
S. Baehr
, et al. (543 additional authors not shown)
Abstract:
We present a reconstruction in early data of the semileptonic decay $B^+ \to π^0 \ell^+ ν_\ell$, and first results of a reconstruction of the decays $B^+ \to ρ^0 \ell^+ ν_\ell$ and $B^0 \to ρ^- \ell^+ ν_\ell$ in a sample corresponding to 62.8 fb$^{-1}$ of Belle II data using hadronic $B$-tagging via the full-event-interpretation algorithm. We determine the total branching fractions via fits to the…
▽ More
We present a reconstruction in early data of the semileptonic decay $B^+ \to π^0 \ell^+ ν_\ell$, and first results of a reconstruction of the decays $B^+ \to ρ^0 \ell^+ ν_\ell$ and $B^0 \to ρ^- \ell^+ ν_\ell$ in a sample corresponding to 62.8 fb$^{-1}$ of Belle II data using hadronic $B$-tagging via the full-event-interpretation algorithm. We determine the total branching fractions via fits to the distribution of the square of the missing mass, and find $\mathcal{B}(B^+ \to π^0 \ell^+ ν_\ell)$ = (8.29 $\pm$ 1.99(stat) $\pm$ 0.46(syst)) $\times 10^{-5}$. We obtain $95\%$ CL upper limits on the branching fractions with $\mathcal{B}(B^0 \to ρ^- \ell^+ ν_\ell) < 3.37 \times 10^{-4}$ and $\mathcal{B}(B^+ \to ρ^0 \ell^+ ν_\ell) < 19.7 \times 10^{-5}$. We also obtain an updated branching fraction for the $B^0 \to π^- \ell^+ ν_\ell$ decay, $\mathcal{B}(B^0 \to π^- \ell^+ ν_\ell)$ = (1.47 $\pm$ 0.29(stat) $\pm$ 0.05(syst)) $\times 10^{-4}$, based on the sum of the partial branching fractions in three bins of the squared momentum transfer to the leptonic system.
△ Less
Submitted 1 November, 2021;
originally announced November 2021.
-
Measurements of the branching fractions for $B \to K^{*}γ$ decays at Belle II
Authors:
F. Abudinén,
I. Adachi,
R. Adak,
K. Adamczyk,
L. Aggarwal,
P. Ahlburg,
H. Ahmed,
J. K. Ahn,
H. Aihara,
N. Akopov,
A. Aloisio,
F. Ameli,
L. Andricek,
N. Anh Ky,
D. M. Asner,
H. Atmacan,
V. Aulchenko,
T. Aushev,
V. Aushev,
T. Aziz,
V. Babu,
S. Bacher,
H. Bae,
S. Baehr,
S. Bahinipati
, et al. (543 additional authors not shown)
Abstract:
This paper reports a study of $B \to K^{*}γ$ decays using $62.8\pm 0.6$ fb$^{-1}$ of data collected during 2019--2020 by the Belle II experiment at the SuperKEKB $e^{+}e^{-}$ asymmetric-energy collider, corresponding to $(68.2 \pm 0.8) \times 10^6$ $B\overline{B}$ events. We find $454 \pm 28$, $50 \pm 10$, $169 \pm 18$, and $160 \pm 17$ signal events in the decay modes…
▽ More
This paper reports a study of $B \to K^{*}γ$ decays using $62.8\pm 0.6$ fb$^{-1}$ of data collected during 2019--2020 by the Belle II experiment at the SuperKEKB $e^{+}e^{-}$ asymmetric-energy collider, corresponding to $(68.2 \pm 0.8) \times 10^6$ $B\overline{B}$ events. We find $454 \pm 28$, $50 \pm 10$, $169 \pm 18$, and $160 \pm 17$ signal events in the decay modes $B^{0} \to K^{*0}[K^{+}π^{-}]γ$, $B^{0} \to K^{*0}[K^0_{\rm S}π^{0}]γ$, $B^{+} \to K^{*+}[K^{+}π^{0}]γ$, and $B^{+} \to K^{*+}[K^{+}π^{0}]γ$, respectively. The uncertainties quoted for the signal yield are statistical only. We report the branching fractions of these decays: $$\mathcal{B} [B^{0} \to K^{*0}[K^{+}π^{-}]γ] = (4.5 \pm 0.3 \pm 0.2) \times 10^{-5}, $$ $$\mathcal{B} [B^{0} \to K^{*0}[K^0_{\rm S}π^{0}]γ] = (4.4 \pm 0.9 \pm 0.6) \times 10^{-5},$$ $$\mathcal{B} [B^{+} \to K^{*+}[K^{+}π^{0}]γ] = (5.0 \pm 0.5 \pm 0.4)\times 10^{-5},\text{ and}$$ $$\mathcal{B} [B^{+} \to K^{*+}[K^0_{\rm S}π^{+}]γ] = (5.4 \pm 0.6 \pm 0.4) \times 10^{-5},$$ where the first uncertainty is statistical, and the second is systematic. The results are consistent with world-average values.
△ Less
Submitted 15 October, 2021;
originally announced October 2021.
-
Angular analysis of $B^+ \to ρ^+ρ^0$ decays reconstructed in 2019-2020 Belle II data
Authors:
Belle II collaboration,
F. Abudinén,
I. Adachi,
R. Adak,
K. Adamczyk,
P. Ahlburg,
J. K. Ahn,
H. Aihara,
N. Akopov,
A. Aloisio,
F. Ameli,
L. Andricek,
N. Anh Ky,
D. M. Asner,
H. Atmacan,
V. Aulchenko,
T. Aushev,
V. Aushev,
T. Aziz,
V. Babu,
S. Bacher,
S. Baehr,
S. Bahinipati,
A. M. Bakich,
P. Bambade
, et al. (527 additional authors not shown)
Abstract:
We report on the first Belle II measurement of the branching fraction ($\mathcal{B}$) and longitudinal polarization fraction ($f_L$) of $B^+\to ρ^+ρ^0$ decays. We reconstruct $B^+\to ρ^+(\to π^+π^0(\to γγ))ρ^0(\to π^+π^-)$ decays in a sample of SuperKEKB electron-positron collisions collected by the Belle II experiment in 2019 and 2020 at the $Υ$(4S) resonance and corresponding to $62.8$ fb…
▽ More
We report on the first Belle II measurement of the branching fraction ($\mathcal{B}$) and longitudinal polarization fraction ($f_L$) of $B^+\to ρ^+ρ^0$ decays. We reconstruct $B^+\to ρ^+(\to π^+π^0(\to γγ))ρ^0(\to π^+π^-)$ decays in a sample of SuperKEKB electron-positron collisions collected by the Belle II experiment in 2019 and 2020 at the $Υ$(4S) resonance and corresponding to $62.8$ fb$^{-1}$ of integrated luminosity. We fit the distributions of the difference between expected and observed $B$ candidate energy, continuum-suppression variable, dipion masses, and angular distributions of the resulting samples, to determine a signal yield of $104\pm16$ events. The signal yields are corrected for efficiencies determined from simulation and control data samples to obtain $\mathcal{B}(B^+ \to ρ^+ρ^0) = [20.6 \pm 3.2(\rm stat) \pm 4.0(\rm syst)]\times 10^{-6}$, and $f_L(B^+ \to ρ^+ρ^0) = 0.936 ^{+0.049}_{-0.041}(\rm stat)\pm 0.021(\rm syst)$. This first Belle II $B^+ \to ρ^+ρ^0$ angular analysis yields results compatible with previous determinations, and indicates Belle II performance superior to early Belle results.
△ Less
Submitted 28 September, 2021; v1 submitted 23 September, 2021;
originally announced September 2021.
-
Measurements of branching fractions and CP-violating charge asymmetries in multibody charmless $B$ decays reconstructed in 2019-2020 Belle II data
Authors:
Belle II collaboration,
F. Abudinén,
I. Adachi,
R. Adak,
K. Adamczyk,
P. Ahlburg,
J. K. Ahn,
H. Aihara,
N. Akopov,
A. Aloisio,
F. Ameli,
L. Andricek,
N. Anh Ky,
D. M. Asner,
H. Atmacan,
V. Aulchenko,
T. Aushev,
V. Aushev,
T. Aziz,
V. Babu,
S. Bacher,
S. Baehr,
S. Bahinipati,
A. M. Bakich,
P. Bambade
, et al. (527 additional authors not shown)
Abstract:
We report on measurements of branching fractions ($\mathcal{B}$) and CP-violating charge asymmetries ($\mathcal{A}_{\rm CP}$) of multibody charmless $B$ decays reconstructed by the Belle II experiment at the SuperKEKB electron-positron collider. We use a sample of collisions collected in 2019 and 2020 at the $Υ(4S)$ resonance and corresponding to $62.8$ fb$^{-1}$ of integrated luminosity. We use s…
▽ More
We report on measurements of branching fractions ($\mathcal{B}$) and CP-violating charge asymmetries ($\mathcal{A}_{\rm CP}$) of multibody charmless $B$ decays reconstructed by the Belle II experiment at the SuperKEKB electron-positron collider. We use a sample of collisions collected in 2019 and 2020 at the $Υ(4S)$ resonance and corresponding to $62.8$ fb$^{-1}$ of integrated luminosity. We use simulation to determine optimized event selections. The $ΔE$ and $M_{\rm bc}$ distributions of the resulting samples are fit to determine signal yields of approximately 690, 840, and 380 decays for the channels $B^+ \to K^+K^-K^+$, $B^+ \to K^+π^-π^+$, and $B^0 \to K^+π^-π^0$, respectively. These yields are corrected for efficiencies determined from simulation and control data samples to obtain $\mathcal{B}(B^+ \to K^+K^-K^+) = [35.8 \pm 1.6(\rm stat) \pm 1.4 (\rm syst)]\times 10^{-6}$, $\mathcal{B}(B^+ \to K^+π^-π^+) = [67.0 \pm 3.3 (\rm stat)\pm 2.3 (\rm syst)]\times 10^{-6}$, $\mathcal{B}(B^0 \to K^+π^-π^0) = [38.1 \pm 3.5 (\rm stat)\pm 3.9 (\rm syst)]\times 10^{-6}$, $\mathcal{A}_{\rm CP}(B^+ \to K^+K^-K^+) = -0.103 \pm 0.042(\rm stat) \pm 0.020 (\rm syst)$, $\mathcal{A}_{\rm CP}(B^+ \to K^+π^-π^+) = -0.010 \pm 0.050 (\rm stat)\pm 0.021(\rm syst)$, and $\mathcal{A}_{\rm CP}(B^0 \to K^+π^-π^0) = 0.207 \pm 0.088 (\rm stat)\pm 0.011(\rm syst)$. Results are consistent with previous measurements and demonstrate detector performance comparable with the best Belle results.
△ Less
Submitted 28 September, 2021; v1 submitted 22 September, 2021;
originally announced September 2021.
-
Precise measurement of the $D^0$ and $D^+$ lifetimes at Belle II
Authors:
Belle II collaboration,
F. Abudinén,
I. Adachi,
K. Adamczyk,
L. Aggarwal,
H. Ahmed,
H. Aihara,
N. Akopov,
A. Aloisio,
N. Anh Ky,
D. M. Asner,
H. Atmacan,
V. Aushev,
V. Babu,
S. Bacher,
H. Bae,
S. Baehr,
S. Bahinipati,
P. Bambade,
Sw. Banerjee,
S. Bansal,
M. Barrett,
J. Baudot,
M. Bauer,
A. Baur
, et al. (363 additional authors not shown)
Abstract:
We report a measurement of the $D^0$ and $D^+$ lifetimes using $D^0\to K^-π^+$ and $D^+\to K^-π^+π^+$ decays reconstructed in $e^+e^-\to c\bar{c}$ data recorded by the Belle II experiment at the SuperKEKB asymmetric-energy $e^+e^-$ collider. The data, collected at center-of-mass energies at or near the $Υ(4S)$ resonance, correspond to an integrated luminosity of $72\,{\rm fb}^{-1}$. The results,…
▽ More
We report a measurement of the $D^0$ and $D^+$ lifetimes using $D^0\to K^-π^+$ and $D^+\to K^-π^+π^+$ decays reconstructed in $e^+e^-\to c\bar{c}$ data recorded by the Belle II experiment at the SuperKEKB asymmetric-energy $e^+e^-$ collider. The data, collected at center-of-mass energies at or near the $Υ(4S)$ resonance, correspond to an integrated luminosity of $72\,{\rm fb}^{-1}$. The results, $τ(D^0) = 410.5\pm1.1\,{\rm(stat)}\pm0.8\,{\rm(syst.)}\,{\rm fs}$ and $τ(D^+) = 1030.4\pm4.7\,{\rm(stat)}\pm3.1\,{\rm(syst.)}\,{\rm fs}$, are the most precise to date and are consistent with previous determinations.
△ Less
Submitted 19 November, 2021; v1 submitted 6 August, 2021;
originally announced August 2021.
-
Effect of Input Size on the Classification of Lung Nodules Using Convolutional Neural Networks
Authors:
Gorkem Polat,
Yesim Dogrusoz Serinagaoglu,
Ugur Halici
Abstract:
Recent studies have shown that lung cancer screening using annual low-dose computed tomography (CT) reduces lung cancer mortality by 20% compared to traditional chest radiography. Therefore, CT lung screening has started to be used widely all across the world. However, analyzing these images is a serious burden for radiologists. The number of slices in a CT scan can be up to 600. Therefore, comput…
▽ More
Recent studies have shown that lung cancer screening using annual low-dose computed tomography (CT) reduces lung cancer mortality by 20% compared to traditional chest radiography. Therefore, CT lung screening has started to be used widely all across the world. However, analyzing these images is a serious burden for radiologists. The number of slices in a CT scan can be up to 600. Therefore, computer-aided-detection (CAD) systems are very important for faster and more accurate assessment of the data. In this study, we proposed a framework that analyzes CT lung screenings using convolutional neural networks (CNNs) to reduce false positives. We trained our model with different volume sizes and showed that volume size plays a critical role in the performance of the system. We also used different fusions in order to show their power and effect on the overall accuracy. 3D CNNs were preferred over 2D CNNs because 2D convolutional operations applied to 3D data could result in information loss. The proposed framework has been tested on the dataset provided by the LUNA16 Challenge and resulted in a sensitivity of 0.831 at 1 false positive per scan.
△ Less
Submitted 11 July, 2021;
originally announced July 2021.
-
Measurement of the branching fraction for $B^{0} \rightarrow π^{0} π^{0}$ decays reconstructed in 2019-2020 Belle II data
Authors:
Belle II Collaboration,
F. Abudinén,
I. Adachi,
R. Adak,
K. Adamczyk,
P. Ahlburg,
J. K. Ahn,
H. Aihara,
N. Akopov,
A. Aloisio,
F. Ameli,
L. Andricek,
N. Anh Ky,
D. M. Asner,
H. Atmacan,
V. Aulchenko,
T. Aushev,
V. Aushev,
T. Aziz,
V. Babu,
S. Bacher,
S. Baehr,
S. Bahinipati,
A. M. Bakich,
P. Bambade
, et al. (529 additional authors not shown)
Abstract:
We report the first reconstruction of the $B^{0} \to π^{0} π^{0}$ decay mode at Belle II using samples of 2019 and 2020 data that correspond to 62.8 fb$^{-1}$ of integrated luminosity. We find $14.0^{+6.8}_{-5.6}$ signal decays, corresponding to a significance of 3.4 standard deviations and determine a branching ratio of…
▽ More
We report the first reconstruction of the $B^{0} \to π^{0} π^{0}$ decay mode at Belle II using samples of 2019 and 2020 data that correspond to 62.8 fb$^{-1}$ of integrated luminosity. We find $14.0^{+6.8}_{-5.6}$ signal decays, corresponding to a significance of 3.4 standard deviations and determine a branching ratio of $\mathcal{B}(B^{0} \rightarrow π^{0} π^{0}) = [0.98^{+0.48}_{-0.39} \pm 0.27] \times 10^{-6}$. The results agree with previous determinations and contribute important information to an early assessment of detector performance and Belle II's potential for future determinations of $α/φ_2$ using $B \rightarrow ππ$ modes.
△ Less
Submitted 28 September, 2021; v1 submitted 5 July, 2021;
originally announced July 2021.
-
Rediscovery of $B^0\to J\mskip 1mu / ψ\mskip 2mu K^0_{\scriptscriptstyle L}$ at Belle II
Authors:
Belle II Collaboration,
F. Abudinén,
I. Adachi,
R. Adak,
K. Adamczyk,
P. Ahlburg,
J. K. Ahn,
H. Aihara,
N. Akopov,
A. Aloisio,
F. Ameli,
L. Andricek,
N. Anh Ky,
D. M. Asner,
H. Atmacan,
V. Aulchenko,
T. Aushev,
V. Aushev,
T. Aziz,
V. Babu,
S. Bacher,
S. Baehr,
S. Bahinipati,
A. M. Bakich,
P. Bambade
, et al. (523 additional authors not shown)
Abstract:
We present preliminary results on the reconstruction of the $B^0\to J\mskip 1mu / ψ\mskip 2mu K^0_{\scriptscriptstyle L}$ decay, where $J\mskip 1mu / ψ\mskip 2mu\toμ^+μ^-$ or $e^+e^-$. Using a dataset corresponding to a luminosity of $62.8\pm0.6\mbox{fb}^{-1}$ collected by the Belle II experiment at the SuperKEKB asymmetric energy $e^+e^-$ collider, we measure a total of $267\pm21$ candidates with…
▽ More
We present preliminary results on the reconstruction of the $B^0\to J\mskip 1mu / ψ\mskip 2mu K^0_{\scriptscriptstyle L}$ decay, where $J\mskip 1mu / ψ\mskip 2mu\toμ^+μ^-$ or $e^+e^-$. Using a dataset corresponding to a luminosity of $62.8\pm0.6\mbox{fb}^{-1}$ collected by the Belle II experiment at the SuperKEKB asymmetric energy $e^+e^-$ collider, we measure a total of $267\pm21$ candidates with $J\mskip 1mu / ψ\mskip 2mu\toμ^+μ^-$ and $226\pm20$ with with $J\mskip 1mu / ψ\mskip 2mu\to e^+e^-$. The quoted errors are statistical only.
△ Less
Submitted 25 June, 2021;
originally announced June 2021.
-
Ad Text Classification with Transformer-Based Natural Language Processing Methods
Authors:
Umut Özdil,
Büşra Arslan,
D. Emre Taşar,
Gökçe Polat,
Şükrü Ozan
Abstract:
In this study, a natural language processing-based (NLP-based) method is proposed for the sector-wise automatic classification of ad texts created on online advertising platforms. Our data set consists of approximately 21,000 labeled advertising texts from 12 different sectors. In the study, the Bidirectional Encoder Representations from Transformers (BERT) model, which is a transformer-based lang…
▽ More
In this study, a natural language processing-based (NLP-based) method is proposed for the sector-wise automatic classification of ad texts created on online advertising platforms. Our data set consists of approximately 21,000 labeled advertising texts from 12 different sectors. In the study, the Bidirectional Encoder Representations from Transformers (BERT) model, which is a transformer-based language model that is recently used in fields such as text classification in the natural language processing literature, was used. The classification efficiencies obtained using a pre-trained BERT model for the Turkish language are shown in detail.
△ Less
Submitted 23 June, 2021; v1 submitted 21 June, 2021;
originally announced June 2021.
-
Measurements of branching fractions and direct CP asymmetries in $B^{0}\to K^{+} π^{-}$, $B^+ \to K_{\rm S}^0π^+$ and $B^0 \to π^+π^-$ using 2019 and 2020 data
Authors:
Belle II Collaboration,
F. Abudinén,
I. Adachi,
R. Adak,
K. Adamczyk,
P. Ahlburg,
J. K. Ahn,
H. Aihara,
N. Akopov,
A. Aloisio,
F. Ameli,
L. Andricek,
N. Anh Ky,
D. M. Asner,
H. Atmacan,
V. Aulchenko,
T. Aushev,
V. Aushev,
T. Aziz,
V. Babu,
S. Bacher,
S. Baehr,
S. Bahinipati,
A. M. Bakich,
P. Bambade
, et al. (527 additional authors not shown)
Abstract:
We report updated measurements of branching fractions ($\mathcal{B}$) and CP-violating charge asymmetries ($\mathcal{A_{\rm CP}}$) for charmless $B$ decays at Belle II, which operates on or near the $Υ$(4S) resonance at the SuperKEKB asymmetric energy $e^{+}e^{-}$ collider. We use samples of 2019 and 2020 data corresponding to 62.8 fb$^{-1}$ of integrated luminosity. The samples are analysed using…
▽ More
We report updated measurements of branching fractions ($\mathcal{B}$) and CP-violating charge asymmetries ($\mathcal{A_{\rm CP}}$) for charmless $B$ decays at Belle II, which operates on or near the $Υ$(4S) resonance at the SuperKEKB asymmetric energy $e^{+}e^{-}$ collider. We use samples of 2019 and 2020 data corresponding to 62.8 fb$^{-1}$ of integrated luminosity. The samples are analysed using two-dimensional fits in $ΔE$ and $M_{\it bc}$ to determine signal yields of approximately 568, 103, and 115 decays for the channels $B^0 \to K^+π^-$, $B^+ \to K_{\rm S}^0π^+$, and $B^0 \to π^+π^-$, respectively. Signal yields are corrected for efficiencies determined from simulation and control data samples to obtain branching fractions and CP-violating asymmetries for flavour-specific channels. The results are compatible with known determinations and contribute important information to an early assessment of Belle II detector performance.
△ Less
Submitted 7 June, 2021;
originally announced June 2021.
-
Measurement of the time-integrated mixing probability $χ_d$ with a semileptonic double-tagging strategy and $34.6 {\rm fb}^{-1}$ of Belle II collision data
Authors:
Belle II Collaboration,
F. Abudinén,
I. Adachi,
R. Adak,
K. Adamczyk,
P. Ahlburg,
J. K. Ahn,
H. Aihara,
N. Akopov,
A. Aloisio,
F. Ameli,
L. Andricek,
N. Anh Ky,
D. M. Asner,
H. Atmacan,
V. Aulchenko,
T. Aushev,
V. Aushev,
T. Aziz,
V. Babu,
S. Bacher,
S. Baehr,
S. Bahinipati,
A. M. Bakich,
P. Bambade
, et al. (528 additional authors not shown)
Abstract:
We present the first measurement of the time-integrated mixing probability $χ_d$ using Belle II data collected at a center-of-mass (CM) energy of 10.58 GeV, corresponding to the mass of the $Υ$(4S) resonance, with an integrated luminosity of $34.6 {\rm fb}^{-1}$ at the SuperKEKB $e^+ e^-$ collider. We reconstruct pairs of B mesons both of which decay to semileptonic final states. Using a novel met…
▽ More
We present the first measurement of the time-integrated mixing probability $χ_d$ using Belle II data collected at a center-of-mass (CM) energy of 10.58 GeV, corresponding to the mass of the $Υ$(4S) resonance, with an integrated luminosity of $34.6 {\rm fb}^{-1}$ at the SuperKEKB $e^+ e^-$ collider. We reconstruct pairs of B mesons both of which decay to semileptonic final states. Using a novel methodology, we measure $χ_d = 0.187 \pm 0.010 \text{ (stat.)} \pm 0.019 \text{ (syst.)}$, which is compatible with existing indirect and direct determinations.
△ Less
Submitted 1 June, 2021;
originally announced June 2021.
-
Measurements of branching fractions and direct ${\it CP}$-violating asymmetries in $B^+ \to K^+ π^0~\mbox{and}~π^+ π^0$ decays using 2019 and 2020 Belle II data
Authors:
F. Abudinén,
I. Adachi,
R. Adak,
K. Adamczyk,
P. Ahlburg,
J. K. Ahn,
H. Aihara,
N. Akopov,
A. Aloisio,
F. Ameli,
L. Andricek,
N. Anh Ky,
D. M. Asner,
H. Atmacan,
V. Aulchenko,
T. Aushev,
V. Aushev,
T. Aziz,
V. Babu,
S. Bacher,
S. Baehr,
S. Bahinipati,
A. M. Bakich,
P. Bambade,
Sw. Banerjee
, et al. (527 additional authors not shown)
Abstract:
We report measurements of branching fractions ($\mathcal B$) and direct ${\it CP}$-violating asymmetries ($\mathcal A_{\it CP}$) for the decays $B^+\to K^+π^0$ and $B^+ \to π^+π^0$ reconstructed with the Belle II detector in a sample of asymmetric-energy electron-positron collisions at the $Υ(4S)$ resonance corresponding to 62.8 $\text{fb}^{-1}$ of integrated luminosity. The results are…
▽ More
We report measurements of branching fractions ($\mathcal B$) and direct ${\it CP}$-violating asymmetries ($\mathcal A_{\it CP}$) for the decays $B^+\to K^+π^0$ and $B^+ \to π^+π^0$ reconstructed with the Belle II detector in a sample of asymmetric-energy electron-positron collisions at the $Υ(4S)$ resonance corresponding to 62.8 $\text{fb}^{-1}$ of integrated luminosity. The results are $\mathcal{B}(B^+ \to K^+π^0) = [11.9 ^{+1.1}_{-1.0} (\rm stat)\pm 1.6(\rm syst)]\times 10^{-6}$, $\mathcal{B}(B^+ \to π^+π^0) = [5.5 ^{+1.0}_{-0.9} (\rm stat)\pm 0.7(\rm syst)]\times 10^{-6}$, $\mathcal A_{\it CP}(B^+ \to K^+π^0) = -0.09 \pm 0.09 (\rm stat)\pm 0.03(\rm syst)$, and $\mathcal A_{\it CP}(B^+ \to π^+π^0) = -0.04 \pm 0.17 (\rm stat)\pm 0.06(\rm syst)$. The results are consistent with previous measurements and show a detector performance comparable with early Belle performance.
△ Less
Submitted 10 May, 2021;
originally announced May 2021.
-
First search for direct $CP$-violating asymmetry in $B^0 \to K^0 π^0$ decays at Belle II
Authors:
Belle II Collaboration,
F. Abudinén,
I. Adachi,
R. Adak,
K. Adamczyk,
P. Ahlburg,
J. K. Ahn,
H. Aihara,
N. Akopov,
A. Aloisio,
F. Ameli,
L. Andricek,
N. Anh Ky,
D. M. Asner,
H. Atmacan,
V. Aulchenko,
T. Aushev,
V. Aushev,
T. Aziz,
V. Babu,
S. Bacher,
S. Baehr,
S. Bahinipati,
A. M. Bakich,
P. Bambade
, et al. (529 additional authors not shown)
Abstract:
We report on the first measurement of the direct $CP$-violating asymmetry ($\mathcal{A}$) in the charmless decay $B^0 \to K^0π^0$ at Belle II and an updated measurement of its branching fraction ($\mathcal{B}$). We use a sample of electron-positron collisions collected in 2019 and 2020 at the $Υ(4S)$ resonance and corresponding to $62.8$ $\text{fb}^{-1}$ of integrated luminosity. We reconstruct an…
▽ More
We report on the first measurement of the direct $CP$-violating asymmetry ($\mathcal{A}$) in the charmless decay $B^0 \to K^0π^0$ at Belle II and an updated measurement of its branching fraction ($\mathcal{B}$). We use a sample of electron-positron collisions collected in 2019 and 2020 at the $Υ(4S)$ resonance and corresponding to $62.8$ $\text{fb}^{-1}$ of integrated luminosity. We reconstruct and select about $50$ $B^0 \to K_S^0 π^0$ candidates, and we measure $\mathcal{A}_{K^0π^0} = -0.40_{-0.44}^{+0.46} (\text{stat}) \pm 0.04 (\text{syst})$ and $\mathcal{B}(B^0 \to K^0 π^0) = [8.5_{-1.6}^{+1.7} (\text{stat}) \pm 1.2 (\text{syst})] \times 10^{-6}$. This is the first measurement of $CP$ violation in $B^0 \to K^0π^0$ decays reported by Belle II. The results agree with previous determinations and show a detector performance comparable with the best Belle results.
△ Less
Submitted 3 May, 2021; v1 submitted 30 April, 2021;
originally announced April 2021.
-
Search for $B^{+}\to K^{+}ν\barν$ decays using an inclusive tagging method at Belle II
Authors:
Belle II Collaboration,
F. Abudinén,
I. Adachi,
K. Adamczyk,
P. Ahlburg,
H. Aihara,
N. Akopov,
A. Aloisio,
N. Anh Ky,
D. M. Asner,
H. Atmacan,
T. Aushev,
V. Aushev,
A. Baur,
V. Babu,
S. Baehr,
P. Bambade,
Sw. Banerjee,
S. Bansal,
J. Baudot,
J. Becker,
P. K. Behera,
J. V. Bennett,
E. Bernieri,
F. U. Bernlochner
, et al. (355 additional authors not shown)
Abstract:
A search for the flavor-changing neutral-current decay $B^{+}\to K^{+}ν\barν$ is performed at the Belle II experiment at the SuperKEKB asymmetric energy electron-positron collider. The results are based on a data sample corresponding to an integrated luminosity of $63\,\mbox{fb}^{-1}$ collected at the $Υ{(4S)}$ resonance and a sample of $9\,\mbox{fb}^{-1}$ collected at an energy…
▽ More
A search for the flavor-changing neutral-current decay $B^{+}\to K^{+}ν\barν$ is performed at the Belle II experiment at the SuperKEKB asymmetric energy electron-positron collider. The results are based on a data sample corresponding to an integrated luminosity of $63\,\mbox{fb}^{-1}$ collected at the $Υ{(4S)}$ resonance and a sample of $9\,\mbox{fb}^{-1}$ collected at an energy $60\mathrm{\,Me\kern -0.1em V}$ below the resonance. A novel measurement method is employed, which exploits topological properties of the $B^{+}\to K^{+}ν\barν$ decay that differ from both generic bottom-meson decays and light-quark pair production. This inclusive tagging approach offers a higher signal efficiency compared to previous searches. No significant signal is observed. An upper limit on the branching fraction of $B^{+}\to K^{+}ν\barν$ of $4.1 \times 10^{-5}$ is set at the 90% confidence level.
△ Less
Submitted 29 October, 2021; v1 submitted 26 April, 2021;
originally announced April 2021.
-
Measurement of the branching fractions of $B\toη' K$ decays using 2019/2020 Belle II data
Authors:
Belle II Collaboration,
F. Abudinén,
I. Adachi,
R. Adak,
K. Adamczyk,
P. Ahlburg,
J. K. Ahn,
H. Aihara,
N. Akopov,
A. Aloisio,
F. Ameli,
L. Andricek,
N. Anh Ky,
D. M. Asner,
H. Atmacan,
V. Aulchenko,
T. Aushev,
V. Aushev,
T. Aziz,
V. Babu,
S. Bacher,
S. Baehr,
S. Bahinipati,
A. M. Bakich,
P. Bambade
, et al. (523 additional authors not shown)
Abstract:
This note describes the rediscovery of $B\toη' K$ decays in Belle II data, both in the charged and neutral final state: $B_0\toη' K_S$ and $B^\pm\toη' K^\pm$. The $η'$ is searched for in two decay modes: $η'\toηπ^+π^-$ with $η\toγγ$, and $η'\toργ$. The analysis uses data collected in 2019 and 2020 at the SuperKEKB asymmetric $e^+e^-$ collider, with an integrated luminosity of $62.8~fb^{-1}$, corre…
▽ More
This note describes the rediscovery of $B\toη' K$ decays in Belle II data, both in the charged and neutral final state: $B_0\toη' K_S$ and $B^\pm\toη' K^\pm$. The $η'$ is searched for in two decay modes: $η'\toηπ^+π^-$ with $η\toγγ$, and $η'\toργ$. The analysis uses data collected in 2019 and 2020 at the SuperKEKB asymmetric $e^+e^-$ collider, with an integrated luminosity of $62.8~fb^{-1}$, corresponding to $68.2$ million of $B\bar{B}$ pairs produced. The signal yield is obtained via an unbinned maximum likelihood fit to signal sensitive variables, obtaining branching ratios:
$$\mathcal{B}\left(B^\pm\toη'K^\pm\right) = \left(63.4~^{+3.4}_{-3.3}\,(stat)\,\pm3.2\,(syst)\,\right) \times10^{-6} $$
$$\mathcal{B}\left(B_0\toη'K_S\right) = \left(59.9~^{+5.8}_{-5.5}\,(stat)\,\pm2.9\,(syst)\,\right) \times10^{-6} $$ which are consistent with world average.
△ Less
Submitted 12 May, 2021; v1 submitted 13 April, 2021;
originally announced April 2021.
-
Common Limitations of Image Processing Metrics: A Picture Story
Authors:
Annika Reinke,
Minu D. Tizabi,
Carole H. Sudre,
Matthias Eisenmann,
Tim Rädsch,
Michael Baumgartner,
Laura Acion,
Michela Antonelli,
Tal Arbel,
Spyridon Bakas,
Peter Bankhead,
Arriel Benis,
Matthew Blaschko,
Florian Buettner,
M. Jorge Cardoso,
Jianxu Chen,
Veronika Cheplygina,
Evangelia Christodoulou,
Beth Cimini,
Gary S. Collins,
Sandy Engelhardt,
Keyvan Farahani,
Luciana Ferrer,
Adrian Galdran,
Bram van Ginneken
, et al. (68 additional authors not shown)
Abstract:
While the importance of automatic image analysis is continuously increasing, recent meta-research revealed major flaws with respect to algorithm validation. Performance metrics are particularly key for meaningful, objective, and transparent performance assessment and validation of the used automatic algorithms, but relatively little attention has been given to the practical pitfalls when using spe…
▽ More
While the importance of automatic image analysis is continuously increasing, recent meta-research revealed major flaws with respect to algorithm validation. Performance metrics are particularly key for meaningful, objective, and transparent performance assessment and validation of the used automatic algorithms, but relatively little attention has been given to the practical pitfalls when using specific metrics for a given image analysis task. These are typically related to (1) the disregard of inherent metric properties, such as the behaviour in the presence of class imbalance or small target structures, (2) the disregard of inherent data set properties, such as the non-independence of the test cases, and (3) the disregard of the actual biomedical domain interest that the metrics should reflect. This living dynamically document has the purpose to illustrate important limitations of performance metrics commonly applied in the field of image analysis. In this context, it focuses on biomedical image analysis problems that can be phrased as image-level classification, semantic segmentation, instance segmentation, or object detection task. The current version is based on a Delphi process on metrics conducted by an international consortium of image analysis experts from more than 60 institutions worldwide.
△ Less
Submitted 6 December, 2023; v1 submitted 12 April, 2021;
originally announced April 2021.
-
Study of $B\to D^{(*)}h$ decays using $62.8~\mathrm{fb}^{-1}$ of Belle II data
Authors:
Belle II Collaboration,
F. Abudinen,
I. Adachi,
R. Adak,
K. Adamczyk,
P. Ahlburg,
J. K. Ahn,
H. Aihara,
N. Akopov,
A. Aloisio,
F. Ameli,
L. Andricek,
N. Anh Ky,
D. M. Asner,
H. Atmacan,
V. Aulchenko,
T. Aushev,
V. Aushev,
T. Aziz,
V. Babu,
S. Bacher,
S. Baehr,
S. Bahinipati,
A. M. Bakich,
P. Bambade
, et al. (527 additional authors not shown)
Abstract:
We report measurements related to hadronic $B$ decays to final states that contain charm mesons. The analyses are performed on a $62.8~\mathrm{fb}^{-1}$ data set collected by the Belle II experiment at a center-of-mass energy corresponding to the mass of the $Υ(4S)$ resonance. The measurements reported are for the decay modes $B^-\to D^0 h^-$, $B^{-}\to D^{*0}h^-$, $\bar{B}^{0}\to D^{+} h^{-}$ and…
▽ More
We report measurements related to hadronic $B$ decays to final states that contain charm mesons. The analyses are performed on a $62.8~\mathrm{fb}^{-1}$ data set collected by the Belle II experiment at a center-of-mass energy corresponding to the mass of the $Υ(4S)$ resonance. The measurements reported are for the decay modes $B^-\to D^0 h^-$, $B^{-}\to D^{*0}h^-$, $\bar{B}^{0}\to D^{+} h^{-}$ and $\bar{B}^{0}\to D^{*+} h^{-}$, where $h=π$ or $K$. These modes are either signal or control channels for measurements related to the unitarity triangle angle $γ$ in direct or time-dependent $CP$-violation measurements. The reported observables are the ratios between the $B\to D^{(*)}K$ and $B\to D^{(*)}π$ decay rates, which are found to be in agreement with previous measurements.
△ Less
Submitted 8 April, 2021;
originally announced April 2021.
-
Deep learning for detection and segmentation of artefact and disease instances in gastrointestinal endoscopy
Authors:
Sharib Ali,
Mariia Dmitrieva,
Noha Ghatwary,
Sophia Bano,
Gorkem Polat,
Alptekin Temizel,
Adrian Krenzer,
Amar Hekalo,
Yun Bo Guo,
Bogdan Matuszewski,
Mourad Gridach,
Irina Voiculescu,
Vishnusai Yoganand,
Arnav Chavan,
Aryan Raj,
Nhan T. Nguyen,
Dat Q. Tran,
Le Duy Huynh,
Nicolas Boutry,
Shahadate Rezvy,
Haijian Chen,
Yoon Ho Choi,
Anand Subramanian,
Velmurugan Balasubramanian,
Xiaohong W. Gao
, et al. (12 additional authors not shown)
Abstract:
The Endoscopy Computer Vision Challenge (EndoCV) is a crowd-sourcing initiative to address eminent problems in develo** reliable computer aided detection and diagnosis endoscopy systems and suggest a pathway for clinical translation of technologies. Whilst endoscopy is a widely used diagnostic and treatment tool for hollow-organs, there are several core challenges often faced by endoscopists, ma…
▽ More
The Endoscopy Computer Vision Challenge (EndoCV) is a crowd-sourcing initiative to address eminent problems in develo** reliable computer aided detection and diagnosis endoscopy systems and suggest a pathway for clinical translation of technologies. Whilst endoscopy is a widely used diagnostic and treatment tool for hollow-organs, there are several core challenges often faced by endoscopists, mainly: 1) presence of multi-class artefacts that hinder their visual interpretation, and 2) difficulty in identifying subtle precancerous precursors and cancer abnormalities. Artefacts often affect the robustness of deep learning methods applied to the gastrointestinal tract organs as they can be confused with tissue of interest. EndoCV2020 challenges are designed to address research questions in these remits. In this paper, we present a summary of methods developed by the top 17 teams and provide an objective comparison of state-of-the-art methods and methods designed by the participants for two sub-challenges: i) artefact detection and segmentation (EAD2020), and ii) disease detection and segmentation (EDD2020). Multi-center, multi-organ, multi-class, and multi-modal clinical endoscopy datasets were compiled for both EAD2020 and EDD2020 sub-challenges. The out-of-sample generalization ability of detection algorithms was also evaluated. Whilst most teams focused on accuracy improvements, only a few methods hold credibility for clinical usability. The best performing teams provided solutions to tackle class imbalance, and variabilities in size, origin, modality and occurrences by exploring data augmentation, data fusion, and optimal class thresholding techniques.
△ Less
Submitted 17 February, 2021; v1 submitted 12 October, 2020;
originally announced October 2020.
-
Measurements of branching fractions and CP-violating charge asymmetries in charmless $B$ decays reconstructed in 2019--2020 Belle~II data
Authors:
Belle II Collaboration,
F. Abudinén,
I. Adachi,
R. Adak,
K. Adamczyk,
P. Ahlburg,
J. K. Ahn,
H. Aihara,
N. Akopov,
A. Aloisio,
F. Ameli,
L. Andricek,
N. Anh Ky,
D. M. Asner,
H. Atmacan,
V. Aulchenko,
T. Aushev,
V. Aushev,
T. Aziz,
V. Babu,
S. Bacher,
S. Baehr,
S. Bahinipati,
A. M. Bakich,
P. Bambade
, et al. (522 additional authors not shown)
Abstract:
We report on first measurements of branching fractions~($\mathcal{B}$) and CP-violating charge asymmetries~($\mathcal{A}$) in charmless $B$ decays at Belle~II. We use a sample of electron-positron collisions collected in 2019 and 2020 at the $Υ(4S)$ resonance and corresponding to $34.6$\,fb$^{-1}$ of integrated luminosity. We use simulation to determine optimized event selections. The $ΔE$ distrib…
▽ More
We report on first measurements of branching fractions~($\mathcal{B}$) and CP-violating charge asymmetries~($\mathcal{A}$) in charmless $B$ decays at Belle~II. We use a sample of electron-positron collisions collected in 2019 and 2020 at the $Υ(4S)$ resonance and corresponding to $34.6$\,fb$^{-1}$ of integrated luminosity. We use simulation to determine optimized event selections. The $ΔE$ distributions of the resulting samples, restricted in $M_{\rm bc}$, are fit to determine signal yields. Signal yields are corrected for efficiencies determined from simulation and control data samples to obtain branching fractions and CP-violating asymmetries for flavour-specific channels. These are the first measurements in charmless decays reported by Belle~II. Results are compatible with known determinations and show detector performance comparable with the best Belle results offering a reliable basis to assess projections for future reach.
△ Less
Submitted 20 September, 2020;
originally announced September 2020.
-
Measurement of Hadronic Mass Moments $\langle M_X^n \rangle $ in $B \rightarrow X_c \ell ν$ Decays at Belle II
Authors:
Belle II Collaboration,
F. Abudinén,
I. Adachi,
R. Adak,
K. Adamczyk,
P. Ahlburg,
J. K. Ahn,
H. Aihara,
N. Akopov,
A. Aloisio,
F. Ameli,
L. Andricek,
N. Anh Ky,
D. M. Asner,
H. Atmacan,
V. Aulchenko,
T. Aushev,
V. Aushev,
T. Aziz,
V. Babu,
S. Bacher,
S. Baehr,
S. Bahinipati,
A. M. Bakich,
P. Bambade
, et al. (518 additional authors not shown)
Abstract:
We present measurements of the first six hadronic mass moments in semileptonic $B \rightarrow X_c \ell ν$ decays. The hadronic mass moments, together with other observables of inclusive $B$ decays, can be used to determine the CKM matrix element $|{V_{cb}}|$ and mass of the $b$-quark $m_b$ in the context of Heavy Quark Expansions of QCD. The Belle~II data recorded at the $Υ(4S)$ resonance in 2019…
▽ More
We present measurements of the first six hadronic mass moments in semileptonic $B \rightarrow X_c \ell ν$ decays. The hadronic mass moments, together with other observables of inclusive $B$ decays, can be used to determine the CKM matrix element $|{V_{cb}}|$ and mass of the $b$-quark $m_b$ in the context of Heavy Quark Expansions of QCD. The Belle~II data recorded at the $Υ(4S)$ resonance in 2019 and 2020 (March-July), corresponding to an integrated luminosity of $34.6\;\mathrm{fb}^{-1}$, is used for this measurement. The decay $Υ(4S) \rightarrow B \overline{B}$ is reconstructed by applying the hadronic tagging algorithm provided by the Full Event Interpretation to fully reconstruct one $B$ meson. The second $B$ meson is reconstructed inclusively by selecting a high-momentum lepton. The $X_c$ system is identified by the remaining reconstructed tracks and clusters in the electromagnetic calorimeter. We report preliminary results for the hadronic mass moments $\langle M_X^n \rangle $ with $n=1,\dots,6$, measured as a function of a lower cut on the lepton momentum in the signal $B$ rest frame.
△ Less
Submitted 9 September, 2020;
originally announced September 2020.
-
Measurement of the semileptonic $\bar{B}^0 \to D^{*+} \ell^{-} ν_{\ell}$ branching fraction with fully reconstructed $B$ meson decays and 34.6 fb$^{-1}$of Belle II data
Authors:
Belle II Collaboration,
F. Abudinen,
I. Adachi,
R. Adak,
K. Adamczyk,
P. Ahlburg,
J. K. Ahn,
H. Aihara,
N. Akopov,
A. Aloisio,
F. Ameli,
K. Amirie,
L. Andricek,
N. Anh Ky,
D. M. Asner,
H. Atmacan,
V. Aulchenko,
T. Aushev,
V. Aushev,
T. Aziz,
V. Babu,
S. Bacher,
S. Baehr,
S. Bahinipati,
A. M. Bakich
, et al. (518 additional authors not shown)
Abstract:
We present a first measurement of the $\bar{B^{0}} \rightarrow D^{*+} \ell^{-} ν_{\ell}$ branching fraction using fully reconstructed $B$ meson decays employing the Full Event Interpretation algorithm. Collision events corresponding to an integrated luminosity of \lumi are analyzed, which were recorded by the Belle~II detector operated at the SuperKEKB accelerator complex. We measure…
▽ More
We present a first measurement of the $\bar{B^{0}} \rightarrow D^{*+} \ell^{-} ν_{\ell}$ branching fraction using fully reconstructed $B$ meson decays employing the Full Event Interpretation algorithm. Collision events corresponding to an integrated luminosity of \lumi are analyzed, which were recorded by the Belle~II detector operated at the SuperKEKB accelerator complex. We measure$\cal{B}(\bar{B^{0}} \rightarrow D^{*+} \ell^{-} ν_{\ell}) =4.51 \pm 0.41_{stat}\pm0.27_{syst} \pm0.45_{π_s}$, with the first and second error denoting the statistical and systematic uncertainty, respectively, and the third dominant uncertainty is from the slow pion reconstruction efficiency.
△ Less
Submitted 16 September, 2020; v1 submitted 24 August, 2020;
originally announced August 2020.
-
Exclusive $B^0 \to π^- \ell^+ ν_\ell$ Decays with Hadronic Full Event Interpretation Tagging in 34.6 fb$^{-1}$ of Belle II Data
Authors:
Belle II Collaboration,
F. Abudinén,
I. Adachi,
R. Adak,
K. Adamczyk,
P. Ahlburg,
J. K. Ahn,
H. Aihara,
N. Akopov,
A. Aloisio,
F. Ameli,
L. Andricek,
N. Anh Ky,
D. M. Asner,
H. Atmacan,
V. Aulchenko,
T. Aushev,
V. Aushev,
T. Aziz,
V. Babu,
S. Bacher,
S. Baehr,
S. Bahinipati,
A. M. Bakich,
P. Bambade
, et al. (519 additional authors not shown)
Abstract:
We present the results of the re-discovery of the decay $B^0 \to π^- \ell^+ ν_\ell$ in 34.6 fb$^{-1}$ of Belle II data using hadronic $B$-tagging via the Full Event Interpretation algorithm. We observe 21 signal events on a background of 155 in a fit to the distribution of the square of the missing mass, $M_{\mathrm{miss}}^2$, with a significance of 5.69$σ$, and determine a total branching fractio…
▽ More
We present the results of the re-discovery of the decay $B^0 \to π^- \ell^+ ν_\ell$ in 34.6 fb$^{-1}$ of Belle II data using hadronic $B$-tagging via the Full Event Interpretation algorithm. We observe 21 signal events on a background of 155 in a fit to the distribution of the square of the missing mass, $M_{\mathrm{miss}}^2$, with a significance of 5.69$σ$, and determine a total branching fraction of (1.58 $\pm$ 0.43$_{\mathrm{stat}}$ $\pm$ 0.07$_{\mathrm{sys}}$) $\times 10^{-4}$.
△ Less
Submitted 20 August, 2020;
originally announced August 2020.
-
Studies of the semileptonic $\bar B^0\to D^{*+}\ell^-\barν_\ell$ and $B^-\to D^{0}\ell^-\barν_\ell$ decay processes with 34.6 fb$^{-1}$ of Belle II data
Authors:
Belle II Collaboration,
F. Abudinén,
I. Adachi,
R. Adak,
K. Adamczyk,
P. Ahlburg,
J. K. Ahn,
H. Aihara,
N. Akopov,
A. Aloisio,
F. Ameli,
L. Andricek,
N. Anh Ky,
D. M. Asner,
H. Atmacan,
V. Aulchenko,
T. Aushev,
V. Aushev,
T. Aziz,
V. Babu,
S. Bacher,
S. Baehr,
S. Bahinipati,
A. M. Bakich,
P. Bambade
, et al. (520 additional authors not shown)
Abstract:
We report measurements of the $\bar{B}^0 \to D^{*+} \ell^{-} \barν_l$ and $B^- \to D^{0} \ell^{-} \barν_l$ processes using 34.6 fb$^{-1}$ of collision events recorded by the Belle II experiment at the SuperKEKB asymmetric-energy $e^+ e^-$ collider. For the $B^-\to D^{0}\ell^-\barν_\ell$ channel, we present first studies that isolate this decay from other semileptonic processes and backgrounds. We…
▽ More
We report measurements of the $\bar{B}^0 \to D^{*+} \ell^{-} \barν_l$ and $B^- \to D^{0} \ell^{-} \barν_l$ processes using 34.6 fb$^{-1}$ of collision events recorded by the Belle II experiment at the SuperKEKB asymmetric-energy $e^+ e^-$ collider. For the $B^-\to D^{0}\ell^-\barν_\ell$ channel, we present first studies that isolate this decay from other semileptonic processes and backgrounds. We report a measurement of the $\bar{B}^0 \to D^{*+} \ell^{-} \barν_l$ branching fraction and obtain ${\cal B}(\bar{B}^0 \to D^{*+} \ell^{-} \barν_l) = \left(4.60 \pm 0.05_{\mathrm{stat}}\pm0.17_{\mathrm{syst}} \pm 0.45_{π_s}\right) \%$, in agreement with the world average. Here, the uncertainties are statistical, systematic, and related to slow pion reconstruction, respectively. The systematic uncertainties are limited by the statistics of auxiliary measurements and will improve in the future. We also report differential branching fractions in five bins of the hadronic recoil parameter $w$ for $\bar{B}^0 \to D^{*+} \ell^{-} \barν_l$, unfolded to account for resolution and efficiency effects.
△ Less
Submitted 18 August, 2020; v1 submitted 17 August, 2020;
originally announced August 2020.
-
A calibration of the Belle II hadronic tag-side reconstruction algorithm with $B \rightarrow X\ell ν$ decays
Authors:
Belle II Collaboration,
F. Abudinén,
I. Adachi,
R. Adak,
K. Adamczyk,
P. Ahlburg,
J. K. Ahn,
H. Aihara,
N. Akopov,
A. Aloisio,
F. Ameli,
L. Andricek,
N. Anh Ky,
D. M. Asner,
H. Atmacan,
V. Aulchenko,
T. Aushev,
V. Aushev,
T. Aziz,
V. Babu,
S. Bacher,
S. Baehr,
S. Bahinipati,
A. M. Bakich,
P. Bambade
, et al. (518 additional authors not shown)
Abstract:
Tag-side reconstruction is an important method for reconstructing $B$ meson decays with missing energy. The Belle II tag-side reconstruction algorithm, Full Event Interpretation, relies on a hierarchical reconstruction of $B$ meson decays with multivariate classification employed at each stage of reconstruction. Given the large numbers of classifiers employed and decay chains reconstructed, the pe…
▽ More
Tag-side reconstruction is an important method for reconstructing $B$ meson decays with missing energy. The Belle II tag-side reconstruction algorithm, Full Event Interpretation, relies on a hierarchical reconstruction of $B$ meson decays with multivariate classification employed at each stage of reconstruction. Given the large numbers of classifiers employed and decay chains reconstructed, the performance of the algorithm on data and simulation differs significantly. Here, calibration factors are derived for hadronic tag-side $B$ decays by measuring a signal side decay, $B \rightarrow X\ell ν$, in $34.6$ fb$^{-1}$ of Belle II data. For a very loose selection on the tag-side $B$ multivariate classifier, the calibration factors are $0.65 \pm 0.02$ and $0.83 \pm 0.03$ for tag-side $B^{+}$ and $B^{0}$ mesons, respectively.
△ Less
Submitted 17 August, 2020; v1 submitted 13 August, 2020;
originally announced August 2020.
-
$τ$ lepton mass measurement at Belle II
Authors:
Belle II Collaboration,
F. Abudinén,
I. Adachi,
R. Adak,
K. Adamczyk,
P. Ahlburg,
J. K. Ahn,
H. Aihara,
N. Akopov,
A. Aloisio,
F. Ameli,
L. Andricek,
N. Anh Ky,
D. M. Asner,
H. Atmacan,
V. Aulchenko,
T. Aushev,
V. Aushev,
T. Aziz,
V. Babu,
S. Bacher,
S. Baehr,
S. Bahinipati,
A. M. Bakich,
P. Bambade
, et al. (517 additional authors not shown)
Abstract:
The reconstruction of tau-pair production, $e^{+}e^{-} \to τ^{+}τ^{-}$, from the subsequent 3-prong ($τ^{+} \rightarrow π^{+} π^{-} π^{+} \barν_τ$) and 1-prong ($τ^{-} \to \ell^{-} \barν_{\ell} ν_τ$, $τ^{-} \to h^{-} ν_τ$ or $τ^{-} \to π^{-} π^0 ν_τ$) decays, is presented using 8.8 fb$^{-1}$ of $e^{+}e^{-}$ collision data of Belle II at the center-of-mass energy $\sqrt{s} = m_{Υ(4S)}$. The pseudom…
▽ More
The reconstruction of tau-pair production, $e^{+}e^{-} \to τ^{+}τ^{-}$, from the subsequent 3-prong ($τ^{+} \rightarrow π^{+} π^{-} π^{+} \barν_τ$) and 1-prong ($τ^{-} \to \ell^{-} \barν_{\ell} ν_τ$, $τ^{-} \to h^{-} ν_τ$ or $τ^{-} \to π^{-} π^0 ν_τ$) decays, is presented using 8.8 fb$^{-1}$ of $e^{+}e^{-}$ collision data of Belle II at the center-of-mass energy $\sqrt{s} = m_{Υ(4S)}$. The pseudomass technique developed by the ARGUS experiment is used to measure the $τ$-lepton mass $m_τ$ in the 3-prong $τ^{+} \to π^{+} π^{-} π^{+} \barν_τ $ decay, resulting in $m_τ = 1777.28 \pm 0.75~{\rm (stat.)} \pm 0.33 ~{\rm (sys.)}~{\rm{MeV}/\rm{c}^2}$.
△ Less
Submitted 9 March, 2021; v1 submitted 11 August, 2020;
originally announced August 2020.
-
Rediscovery of $B \to φK^{(*)}$ decays and measurement of the longitudinal polarization fraction $f_L$ in $B \to φK^{*}$ decays using the Summer 2020 Belle II dataset
Authors:
F. Abudinén,
I. Adachi,
R. Adak,
K. Adamczyk,
P. Ahlburg,
J. K. Ahn,
H. Aihara,
N. Akopov,
A. Aloisio,
F. Ameli,
L. Andricek,
N. Anh Ky,
D. M. Asner,
H. Atmacan,
V. Aulchenko,
T. Aushev,
V. Aushev,
T. Aziz,
V. Babu,
S. Bacher,
S. Baehr,
S. Bahinipati,
A. M. Bakich,
P. Bambade,
Sw. Banerjee
, et al. (516 additional authors not shown)
Abstract:
We utilize a sample of 34.6 fb$^{-1}$, collected by the Belle II experiment at the SuperKEKB asymmetric energy $e^+e^-$ collider, to search for the $B^+ \to φK^+$, $B^+ \to φK^{*+}$, $B^0 \to φK^0_S$, and $B^0 \to φK^{*0}$ decays. Charmless hadronic $B$ decays represent an important part of the Belle II physics program, and are an ideal benchmark to test the detector capabilities in terms of track…
▽ More
We utilize a sample of 34.6 fb$^{-1}$, collected by the Belle II experiment at the SuperKEKB asymmetric energy $e^+e^-$ collider, to search for the $B^+ \to φK^+$, $B^+ \to φK^{*+}$, $B^0 \to φK^0_S$, and $B^0 \to φK^{*0}$ decays. Charmless hadronic $B$ decays represent an important part of the Belle II physics program, and are an ideal benchmark to test the detector capabilities in terms of tracking efficiency, charged particle identification, vertexing, and advanced analysis techniques. Each channel is observed with a significance that exceeds 5 standard deviations, and we obtain measurements of their branching ratios that are in good agreement with the world averages. For the $B \to φK^*$ modes, we also perform a measurement of the longitudinal polarization fraction $f_L$.
△ Less
Submitted 9 August, 2020;
originally announced August 2020.
-
First flavor tagging calibration using 2019 Belle II data
Authors:
Belle II Collaboration,
F. Abudinén,
I. Adachi,
R. Adak,
K. Adamczyk,
P. Ahlburg,
J. K. Ahn,
H. Aihara,
N. Akopov,
A. Aloisio,
F. Ameli,
L. Andricek,
N. Anh Ky,
D. M. Asner,
H. Atmacan,
V. Aulchenko,
T. Aushev,
V. Aushev,
T. Aziz,
V. Babu,
S. Bacher,
S. Baehr,
S. Bahinipati,
A. M. Bakich,
P. Bambade
, et al. (518 additional authors not shown)
Abstract:
We report on the first calibration of the standard Belle II $B$-flavor tagger using the full data set collected at the $Υ(4{\rm S})$ resonance in 2019 with the Belle II detector at the SuperKEKB collider, corresponding to 8.7 fb$^{-1}$ of integrated luminosity. The calibration is performed by reconstructing various hadronic charmed $B$-meson decays with flavor-specific final states. We use simulat…
▽ More
We report on the first calibration of the standard Belle II $B$-flavor tagger using the full data set collected at the $Υ(4{\rm S})$ resonance in 2019 with the Belle II detector at the SuperKEKB collider, corresponding to 8.7 fb$^{-1}$ of integrated luminosity. The calibration is performed by reconstructing various hadronic charmed $B$-meson decays with flavor-specific final states. We use simulation to optimize our event selection criteria and to train the flavor tagging algorithm. We determine the tagging efficiency and the fraction of wrongly identified tag-side $B$~candidates from a measurement of the time-integrated $B^0-\overline{B}^0$ mixing probability. The total effective efficiency is measured to be $\varepsilon_{\rm eff} = \big(33.8 \pm 3.6(\text{stat}) \pm 1.6(\text{sys})\big)\%$, which is in good agreement with the predictions from simulation and comparable with the best one obtained by the Belle experiment. The results show a good understanding of the detector performance and offer a basis for future calibrations.
△ Less
Submitted 6 August, 2020;
originally announced August 2020.
-
Search for Axion-Like Particles produced in $e^+e^-$ collisions at Belle II
Authors:
Belle II collaboration,
F. Abudinen,
I. Adachi,
H. Aihara,
N. Akopov,
A. Aloisio,
F. Ameli,
N. Anh Ky,
D. M. Asner,
T. Aushev,
V. Aushev,
V. Babu,
S. Baehr,
S. Bahinipati,
P. Bambade,
Sw. Banerjee,
S. Bansal,
J. Baudot,
J. Becker,
P. K. Behera,
J. V. Bennett,
E. Bernieri,
F. U. Bernlochner,
M. Bertemes,
M. Bessner
, et al. (312 additional authors not shown)
Abstract:
We present a search for the direct production of a light pseudoscalar $a$ decaying into two photons with the Belle II detector at the SuperKEKB collider. We search for the process ${e^+e^-\toγa, a \toγγ}$ in the mass range ${0.2} \,< m_a < {9.7}\,{\text{GeV/$c$}^2}$ using data corresponding to an integrated luminosity of $(445\pm 3)\,\text{pb}^{-1}$. Light pseudoscalars interacting predominantly w…
▽ More
We present a search for the direct production of a light pseudoscalar $a$ decaying into two photons with the Belle II detector at the SuperKEKB collider. We search for the process ${e^+e^-\toγa, a \toγγ}$ in the mass range ${0.2} \,< m_a < {9.7}\,{\text{GeV/$c$}^2}$ using data corresponding to an integrated luminosity of $(445\pm 3)\,\text{pb}^{-1}$. Light pseudoscalars interacting predominantly with standard model gauge bosons (so-called axion-like particles or ALPs) are frequently postulated in extensions of the standard model. We find no evidence for ALPs and set 95% confidence level upper limits on the coupling strength $g_{aγγ}$ of ALPs to photons at the level of $10^{-3}\,{\text{GeV}^{-1}}$. The limits are the most restrictive to date for $0.2\,<\,m_a\,<\,1\,{\text{GeV/$c$}^2}$.
△ Less
Submitted 20 October, 2020; v1 submitted 26 July, 2020;
originally announced July 2020.
-
False Positive Reduction in Lung Computed Tomography Images using Convolutional Neural Networks
Authors:
Gorkem Polat,
Ugur Halici,
Yesim Serinagaoglu Dogrusoz
Abstract:
Recent studies have shown that lung cancer screening using annual low-dose computed tomography (CT) reduces lung cancer mortality by 20% compared to traditional chest radiography. Therefore, CT lung screening has started to be used widely all across the world. However, analyzing these images is a serious burden for radiologists. In this study, we propose a novel and simple framework that analyzes…
▽ More
Recent studies have shown that lung cancer screening using annual low-dose computed tomography (CT) reduces lung cancer mortality by 20% compared to traditional chest radiography. Therefore, CT lung screening has started to be used widely all across the world. However, analyzing these images is a serious burden for radiologists. In this study, we propose a novel and simple framework that analyzes CT lung screenings using convolutional neural networks (CNNs) and reduces false positives. Our framework shows that even non-complex architectures are very powerful to classify 3D nodule data when compared to traditional methods. We also use different fusions in order to show their power and effect on the overall score. 3D CNNs are preferred over 2D CNNs because data are in 3D, and 2D convolutional operations may result in information loss. Mini-batch is used in order to overcome class-imbalance. Proposed framework has been validated according to the LUNA16 challenge evaluation and got score of 0.786, which is the average sensitivity values at seven predefined false positive (FP) points.
△ Less
Submitted 4 November, 2018;
originally announced November 2018.
