-
MAMA-MIA: A Large-Scale Multi-Center Breast Cancer DCE-MRI Benchmark Dataset with Expert Segmentations
Authors:
Lidia Garrucho,
Claire-Anne Reidel,
Kaisar Kushibar,
Smriti Joshi,
Richard Osuala,
Apostolia Tsirikoglou,
Maciej Bobowicz,
Javier del Riego,
Alessandro Catanese,
Katarzyna Gwoździewicz,
Maria-Laura Cosaka,
Pasant M. Abo-Elhoda,
Sara W. Tantawy,
Shorouq S. Sakrana,
Norhan O. Shawky-Abdelfatah,
Amr Muhammad Abdo-Salem,
Androniki Kozana,
Eugen Divjak,
Gordana Ivanac,
Katerina Nikiforaki,
Michail E. Klontzas,
Rosa García-Dosdá,
Meltem Gulsun-Akpinar,
Oğuz Lafcı,
Ritse Mann
, et al. (8 additional authors not shown)
Abstract:
Current research in breast cancer Magnetic Resonance Imaging (MRI), especially with Artificial Intelligence (AI), faces challenges due to the lack of expert segmentations. To address this, we introduce the MAMA-MIA dataset, comprising 1506 multi-center dynamic contrast-enhanced MRI cases with expert segmentations of primary tumors and non-mass enhancement areas. These cases were sourced from four…
▽ More
Current research in breast cancer Magnetic Resonance Imaging (MRI), especially with Artificial Intelligence (AI), faces challenges due to the lack of expert segmentations. To address this, we introduce the MAMA-MIA dataset, comprising 1506 multi-center dynamic contrast-enhanced MRI cases with expert segmentations of primary tumors and non-mass enhancement areas. These cases were sourced from four publicly available collections in The Cancer Imaging Archive (TCIA). Initially, we trained a deep learning model to automatically segment the cases, generating preliminary segmentations that significantly reduced expert segmentation time. Sixteen experts, averaging 9 years of experience in breast cancer, then corrected these segmentations, resulting in the final expert segmentations. Additionally, two radiologists conducted a visual inspection of the automatic segmentations to support future quality control studies. Alongside the expert segmentations, we provide 49 harmonized demographic and clinical variables and the pretrained weights of the well-known nnUNet architecture trained using the DCE-MRI full-images and expert segmentations. This dataset aims to accelerate the development and benchmarking of deep learning models and foster innovation in breast cancer diagnostics and treatment planning.
△ Less
Submitted 19 June, 2024;
originally announced June 2024.
-
High-resolution synthesis of high-density breast mammograms: Application to improved fairness in deep learning based mass detection
Authors:
Lidia Garrucho,
Kaisar Kushibar,
Richard Osuala,
Oliver Diaz,
Alessandro Catanese,
Javier del Riego,
Maciej Bobowicz,
Fredrik Strand,
Laura Igual,
Karim Lekadir
Abstract:
Computer-aided detection systems based on deep learning have shown good performance in breast cancer detection. However, high-density breasts show poorer detection performance since dense tissues can mask or even simulate masses. Therefore, the sensitivity of mammography for breast cancer detection can be reduced by more than 20% in dense breasts. Additionally, extremely dense cases reported an in…
▽ More
Computer-aided detection systems based on deep learning have shown good performance in breast cancer detection. However, high-density breasts show poorer detection performance since dense tissues can mask or even simulate masses. Therefore, the sensitivity of mammography for breast cancer detection can be reduced by more than 20% in dense breasts. Additionally, extremely dense cases reported an increased risk of cancer compared to low-density breasts. This study aims to improve the mass detection performance in highdensity breasts using synthetic high-density full-field digital mammograms (FFDM) as data augmentation during breast mass detection model training. To this end, a total of five cycle-consistent GAN (CycleGAN) models using three FFDM datasets were trained for low-to-high-density image translation in highresolution mammograms. The training images were split by breast density BIRADS categories, being BI-RADS A almost entirely fatty and BI-RADS D extremely dense breasts. Our results showed that the proposed data augmentation technique improved the sensitivity and precision of mass detection in models trained with small datasets and improved the domain generalization of the models trained with large databases. In addition, the clinical realism of the synthetic images was evaluated in a reader study involving two expert radiologists and one surgical oncologist.
△ Less
Submitted 24 January, 2023; v1 submitted 20 September, 2022;
originally announced September 2022.
-
Broadband ultraviolet-visible frequency combs from cascaded high-harmonic generation in quasi-phase-matched waveguides
Authors:
Jay Rutledge,
Anthony Catanese,
Daniel D. Hickstein,
Scott A. Diddams,
Thomas K. Allison,
Abijith S. Kowligy
Abstract:
High-harmonic generation (HHG) provides short-wavelength light that is useful for precision spectroscopy and probing ultrafast dynamics. We report efficient, phase-coherent harmonic generation up to 9th-order (333 nm) in chirped periodically poled lithium niobate waveguides driven by phase-stable $\leq$12-nJ, 100 fs pulses at 3 $μ$m with 100 MHz repetition rate. A mid-infrared to ultraviolet-visib…
▽ More
High-harmonic generation (HHG) provides short-wavelength light that is useful for precision spectroscopy and probing ultrafast dynamics. We report efficient, phase-coherent harmonic generation up to 9th-order (333 nm) in chirped periodically poled lithium niobate waveguides driven by phase-stable $\leq$12-nJ, 100 fs pulses at 3 $μ$m with 100 MHz repetition rate. A mid-infrared to ultraviolet-visible conversion efficiency as high as 10% is observed, amongst an overall 23% conversion of the fundamental to all harmonics. We verify the coherence of the harmonic frequency combs despite the complex highly nonlinear process. Numerical simulations based on a single broadband envelope equation with quadratic nonlinearity give estimates for the conversion efficiency within approximately 1 order of magnitude over a wide range of experimental parameters. From this comparison we identify a dimensionless parameter capturing the competition between three-wave mixing and group-velocity walk-off of the harmonics that governs the cascaded HHG physics. These results can inform cascaded HHG in a range of different platforms.
△ Less
Submitted 13 February, 2021; v1 submitted 9 February, 2021;
originally announced February 2021.
-
Mid-infrared frequency comb with 6.7 W average power based on difference frequency generation
Authors:
Anthony Catanese,
Jay Rutledge,
Myles Silfies,
Xinlong Li,
Henry Timmers,
Abijith S. Kowligy,
Alex Lind,
Scott A. Diddams,
Thomas K. Allison
Abstract:
We report on the development of a high-power mid-infrared frequency comb with 100 MHz repetition rate and 100 fs pulse duration. Difference frequency generation is realized between two branches derived from an Er:fiber comb, amplified separately in Yb:fiber and Er:fiber amplifiers. Average powers of 6.7 W and 14.9 W are generated in the 2.9 $μ$m idler and 1.6 $μ$m signal, respectively. With high a…
▽ More
We report on the development of a high-power mid-infrared frequency comb with 100 MHz repetition rate and 100 fs pulse duration. Difference frequency generation is realized between two branches derived from an Er:fiber comb, amplified separately in Yb:fiber and Er:fiber amplifiers. Average powers of 6.7 W and 14.9 W are generated in the 2.9 $μ$m idler and 1.6 $μ$m signal, respectively. With high average power, excellent beam quality, and passive carrier-envelope phase stabilization, this light source is a promising platform for generating broadband frequency combs in the far infrared, visible, and deep ultraviolet.
△ Less
Submitted 7 December, 2019;
originally announced December 2019.
-
Crawling Facebook for Social Network Analysis Purposes
Authors:
Salvatore A. Catanese,
Pasquale De Meo,
Emilio Ferrara,
Giacomo Fiumara,
Alessandro Provetti
Abstract:
We describe our work in the collection and analysis of massive data describing the connections between participants to online social networks. Alternative approaches to social network data collection are defined and evaluated in practice, against the popular Facebook Web site. Thanks to our ad-hoc, privacy-compliant crawlers, two large samples, comprising millions of connections, have been collect…
▽ More
We describe our work in the collection and analysis of massive data describing the connections between participants to online social networks. Alternative approaches to social network data collection are defined and evaluated in practice, against the popular Facebook Web site. Thanks to our ad-hoc, privacy-compliant crawlers, two large samples, comprising millions of connections, have been collected; the data is anonymous and organized as an undirected graph. We describe a set of tools that we developed to analyze specific properties of such social-network graphs, i.e., among others, degree distribution, centrality measures, scaling laws and distribution of friendship.
△ Less
Submitted 31 May, 2011;
originally announced May 2011.
-
Spectrum of TeV Gamma-Rays from the Crab Nebula
Authors:
D. A. Carter-Lewis,
S. Biller,
P. J. Boyle,
J. H. Buckley,
A. Burdett,
J. Bussons Gordo,
M. A. Catanese,
M. F. Cawley,
D. J. Fegan,
J. P. Finley,
J. A. Gaidos,
A. M. Hillas,
F. Krennrich,
R. C. Lamb,
R. W. Lessard,
C. Masterson,
J. E. McEnery,
G. Mohanty,
J. Quinn,
A. J. Rodgers,
H. J. Rose,
F. W. Samuelson,
G. H. Sembroski,
R. Srinivasan,
T. C. Weekes
, et al. (2 additional authors not shown)
Abstract:
The Crab Nebula has become established as the standard candle for TeV gamma-ray astronomy using the atmospheric Cherenkov technique. No evidence for variability has been seen. The spectrum of gamma rays from the Crab Nebula has been measured in the energy range 500 GeV to 8 TeV at the Whipple Observatory by the atmospheric Cherenkov imaging technique. Two methods of analysis involving independen…
▽ More
The Crab Nebula has become established as the standard candle for TeV gamma-ray astronomy using the atmospheric Cherenkov technique. No evidence for variability has been seen. The spectrum of gamma rays from the Crab Nebula has been measured in the energy range 500 GeV to 8 TeV at the Whipple Observatory by the atmospheric Cherenkov imaging technique. Two methods of analysis involving independent Monte Carlo simulations and two databases of observations (1988-89 and 1995-96) were used and gave close agreement. Using the complete spectrum of the Crab Nebula, the spectrum of relativistic electrons is deduced and the spectrum of the resulting inverse Compton gamma-ray emission is in good agreement with the measured spectrum if the ambient magnetic field is about 25-30 nT.
△ Less
Submitted 15 June, 1997;
originally announced June 1997.
-
Constraints on Gamma-ray Emission from the Galactic Plane at 300 TeV
Authors:
A. Borione,
M. A. Catanese,
M. C. Chantell,
C. E. Covault,
J. W. Cronin,
B. E. Fick,
L. F. Fortson,
J. Fowler,
M. A. K. Glasmacher,
K. D. Green,
D. B. Kieda,
J. Matthews,
B. J. Newport,
D. Nitz,
R. A. Ong,
S. Oser,
D. Sinclair,
J. C. van der Velde
Abstract:
We describe a new search for diffuse ultrahigh energy gamma-ray emission associated with molecular clouds in the galactic disk. The Chicago Air Shower Array (CASA), operating in coincidence with the Michigan muon array (MIA), has recorded over 2.2 x 10^{9} air showers from April 4, 1990 to October 7, 1995. We search for gamma rays based upon the muon content of air showers arriving from the dire…
▽ More
We describe a new search for diffuse ultrahigh energy gamma-ray emission associated with molecular clouds in the galactic disk. The Chicago Air Shower Array (CASA), operating in coincidence with the Michigan muon array (MIA), has recorded over 2.2 x 10^{9} air showers from April 4, 1990 to October 7, 1995. We search for gamma rays based upon the muon content of air showers arriving from the direction of the galactic plane. We find no significant evidence for diffuse gamma-ray emission, and we set an upper limit on the ratio of gamma rays to normal hadronic cosmic rays at less than 2.4 x 10^{-5} at 310 TeV (90% confidence limit) from the galactic plane region: (50 degrees < l < 200 degrees); -5 degrees < b < 5 degrees). This limit places a strong constraint on models for emission from molecular clouds in the galaxy. We rule out significant spectral hardening in the outer galaxy, and conclude that emission from the plane at these energies is likely to be dominated by the decay of neutral pions resulting from cosmic rays interactions with passive target gas molecules.
△ Less
Submitted 10 March, 1997;
originally announced March 1997.
