-
PLUTO: Pathology-Universal Transformer
Authors:
Dinkar Juyal,
Harshith Padigela,
Chintan Shah,
Daniel Shenker,
Natalia Harguindeguy,
Yi Liu,
Blake Martin,
Yibo Zhang,
Michael Nercessian,
Miles Markey,
Isaac Finberg,
Kelsey Luu,
Daniel Borders,
Syed Ashar Javed,
Emma Krause,
Raymond Biju,
Aashish Sood,
Allen Ma,
Jackson Nyman,
John Shamshoian,
Guillaume Chhor,
Darpan Sanghavi,
Marc Thibault,
Limin Yu,
Fedaa Najdawi
, et al. (8 additional authors not shown)
Abstract:
Pathology is the study of microscopic inspection of tissue, and a pathology diagnosis is often the medical gold standard to diagnose disease. Pathology images provide a unique challenge for computer-vision-based analysis: a single pathology Whole Slide Image (WSI) is gigapixel-sized and often contains hundreds of thousands to millions of objects of interest across multiple resolutions. In this wor…
▽ More
Pathology is the study of microscopic inspection of tissue, and a pathology diagnosis is often the medical gold standard to diagnose disease. Pathology images provide a unique challenge for computer-vision-based analysis: a single pathology Whole Slide Image (WSI) is gigapixel-sized and often contains hundreds of thousands to millions of objects of interest across multiple resolutions. In this work, we propose PathoLogy Universal TransfOrmer (PLUTO): a light-weight pathology FM that is pre-trained on a diverse dataset of 195 million image tiles collected from multiple sites and extracts meaningful representations across multiple WSI scales that enable a large variety of downstream pathology tasks. In particular, we design task-specific adaptation heads that utilize PLUTO's output embeddings for tasks which span pathology scales ranging from subcellular to slide-scale, including instance segmentation, tile classification, and slide-level prediction. We compare PLUTO's performance to other state-of-the-art methods on a diverse set of external and internal benchmarks covering multiple biologically relevant tasks, tissue types, resolutions, stains, and scanners. We find that PLUTO matches or outperforms existing task-specific baselines and pathology-specific foundation models, some of which use orders-of-magnitude larger datasets and model sizes when compared to PLUTO. Our findings present a path towards a universal embedding to power pathology image analysis, and motivate further exploration around pathology foundation models in terms of data diversity, architectural improvements, sample efficiency, and practical deployability in real-world applications.
△ Less
Submitted 13 May, 2024;
originally announced May 2024.
-
Improved statistical benchmarking of digital pathology models using pairwise frames evaluation
Authors:
Ylaine Gerardin,
John Shamshoian,
Judy Shen,
Nhat Le,
Jamie Prezioso,
John Abel,
Isaac Finberg,
Daniel Borders,
Raymond Biju,
Michael Nercessian,
Vaed Prasad,
Joseph Lee,
Spencer Wyman,
Sid Gupta,
Abigail Emerson,
Bahar Rahsepar,
Darpan Sanghavi,
Ryan Leung,
Limin Yu,
Archit Khosla,
Amaro Taylor-Weiner
Abstract:
Nested pairwise frames is a method for relative benchmarking of cell or tissue digital pathology models against manual pathologist annotations on a set of sampled patches. At a high level, the method compares agreement between a candidate model and pathologist annotations with agreement among pathologists' annotations. This evaluation framework addresses fundamental issues of data size and annotat…
▽ More
Nested pairwise frames is a method for relative benchmarking of cell or tissue digital pathology models against manual pathologist annotations on a set of sampled patches. At a high level, the method compares agreement between a candidate model and pathologist annotations with agreement among pathologists' annotations. This evaluation framework addresses fundamental issues of data size and annotator variability in using manual pathologist annotations as a source of ground truth for model validation. We implemented nested pairwise frames evaluation for tissue classification, cell classification, and cell count prediction tasks and show results for cell and tissue models deployed on an H&E-stained melanoma dataset.
△ Less
Submitted 7 June, 2023;
originally announced June 2023.
-
Synthetic DOmain-Targeted Augmentation (S-DOTA) Improves Model Generalization in Digital Pathology
Authors:
Sai Chowdary Gullapally,
Yibo Zhang,
Nitin Kumar Mittal,
Deeksha Kartik,
Sandhya Srinivasan,
Kevin Rose,
Daniel Shenker,
Dinkar Juyal,
Harshith Padigela,
Raymond Biju,
Victor Minden,
Chirag Maheshwari,
Marc Thibault,
Zvi Goldstein,
Luke Novak,
Nidhi Chandra,
Justin Lee,
Aaditya Prakash,
Chintan Shah,
John Abel,
Darren Fahy,
Amaro Taylor-Weiner,
Anand Sampat
Abstract:
Machine learning algorithms have the potential to improve patient outcomes in digital pathology. However, generalization of these tools is currently limited by sensitivity to variations in tissue preparation, staining procedures and scanning equipment that lead to domain shift in digitized slides. To overcome this limitation and improve model generalization, we studied the effectiveness of two Syn…
▽ More
Machine learning algorithms have the potential to improve patient outcomes in digital pathology. However, generalization of these tools is currently limited by sensitivity to variations in tissue preparation, staining procedures and scanning equipment that lead to domain shift in digitized slides. To overcome this limitation and improve model generalization, we studied the effectiveness of two Synthetic DOmain-Targeted Augmentation (S-DOTA) methods, namely CycleGAN-enabled Scanner Transform (ST) and targeted Stain Vector Augmentation (SVA), and compared them against the International Color Consortium (ICC) profile-based color calibration (ICC Cal) method and a baseline method using traditional brightness, color and noise augmentations. We evaluated the ability of these techniques to improve model generalization to various tasks and settings: four models, two model types (tissue segmentation and cell classification), two loss functions, six labs, six scanners, and three indications (hepatocellular carcinoma (HCC), nonalcoholic steatohepatitis (NASH), prostate adenocarcinoma). We compared these methods based on the macro-averaged F1 scores on in-distribution (ID) and out-of-distribution (OOD) test sets across multiple domains, and found that S-DOTA methods (i.e., ST and SVA) led to significant improvements over ICC Cal and baseline on OOD data while maintaining comparable performance on ID data. Thus, we demonstrate that S-DOTA may help address generalization due to domain shift in real world applications.
△ Less
Submitted 3 May, 2023;
originally announced May 2023.
-
Optical channel waveguides written by high repetition rate femtosecond laser irradiation in Li-Zn fluoroborate glass
Authors:
Sunil Thomas,
T. Toney Fernandez,
Javier Solis,
P. R. Biju,
N. V. Unnikrishnan
Abstract:
Low loss, optical channel waveguides have been successfully produced by high repetition rate, femtosecond laser inscription in a Li-Zn fluoroborate glass (64.9B2O3 + 25Li2O + 10ZnF2 + 0.1Er2O3). High quality waveguides were produced at 500 kHz, 1 MHz and 2 MHz laser repetition rates, showing a refractive index contrast in the range of 3-6 x 10-3 depending on various fluences. Dependence of experim…
▽ More
Low loss, optical channel waveguides have been successfully produced by high repetition rate, femtosecond laser inscription in a Li-Zn fluoroborate glass (64.9B2O3 + 25Li2O + 10ZnF2 + 0.1Er2O3). High quality waveguides were produced at 500 kHz, 1 MHz and 2 MHz laser repetition rates, showing a refractive index contrast in the range of 3-6 x 10-3 depending on various fluences. Dependence of experimental parameters such as average laser power, pulse repetition rate and writing speed on the properties of fabricated waveguides has been discussed. The comparison of optical and compositional characterization techniques evidences an enrichment of B and Zn in the guiding region, while F migrates to the heat diffused region of the written structure.
△ Less
Submitted 6 March, 2015;
originally announced March 2015.
-
Stability of 248-254^Cf isotopes against alpha and cluster radioactivity
Authors:
K. P. Santhosh,
R. K. Biju
Abstract:
Stability of 248-254^Cf nuclei against alpha and cluster emission is studied within our Coulomb and proximity potential model (CPPM). It is found that these nuclei are stable against light clusters (except alpha particle) and instable against heavy cluster emissions. For heavy cluster emissions the daughter nuclei lead to doubly magic 208^Pb or neighbouring one. The effect of quadrapole and hexade…
▽ More
Stability of 248-254^Cf nuclei against alpha and cluster emission is studied within our Coulomb and proximity potential model (CPPM). It is found that these nuclei are stable against light clusters (except alpha particle) and instable against heavy cluster emissions. For heavy cluster emissions the daughter nuclei lead to doubly magic 208^Pb or neighbouring one. The effect of quadrapole and hexadecapole deformations of parent nuclei, daughter nuclei and emitted cluster on half lives are also studied. The computed alpha decay half life values (with including quadrupole deformation β2) are in close agreement with experimental data. Inclusion of quadrupole deformation reduces the height and width of the barrier (increases the barrier penetrability) and hence half life decreases.
△ Less
Submitted 19 April, 2013;
originally announced April 2013.
-
Cluster formation probability in the trans-tin and trans-lead nuclei
Authors:
K. P. Santhosh,
R. K. Biju,
Sabina Sahadevan
Abstract:
Within our fission model, the Coulomb and proximity potential model (CPPM) cluster formation probabilities are calculated for different clusters ranging from carbon to silicon for the parents in the trans-tin and trans- lead regions. It is found that in trans-tin region the 12^C, 16^O, 20^Ne and 24^Mg clusters have maximum cluster formation probability and lowest half lives as compared to other…
▽ More
Within our fission model, the Coulomb and proximity potential model (CPPM) cluster formation probabilities are calculated for different clusters ranging from carbon to silicon for the parents in the trans-tin and trans- lead regions. It is found that in trans-tin region the 12^C, 16^O, 20^Ne and 24^Mg clusters have maximum cluster formation probability and lowest half lives as compared to other clusters. In trans-lead region the 14^C, 18, 20^O, 23^F, 24,26^Ne, 28,30^Mg and 34^Si clusters have the maximum cluster formation probability and minimum half life, which show that alpha like clusters are most probable for emission from trans-tin region while non-alpha clusters are probable from trans-lead region. These results stress the role of neutron proton symmetry and asymmetry of daughter nuclei in these two cases.
△ Less
Submitted 10 May, 2010;
originally announced May 2010.
