1
|
Khene ZE, Kammerer-Jacquet SF, Bigot P, Rabilloud N, Albiges L, Margulis V, De Crevoisier R, Acosta O, Rioux-Leclercq N, Lotan Y, Rouprêt M, Bensalah K. Clinical Application of Digital and Computational Pathology in Renal Cell Carcinoma: A Systematic Review. Eur Urol Oncol 2024; 7:401-411. [PMID: 37925349 DOI: 10.1016/j.euo.2023.10.018] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/29/2023] [Revised: 09/26/2023] [Accepted: 10/24/2023] [Indexed: 11/06/2023]
Abstract
CONTEXT Computational pathology is a new interdisciplinary field that combines traditional pathology with modern technologies such as digital imaging and machine learning to better understand the diagnosis, prognosis, and natural history of many diseases. OBJECTIVE To provide an overview of digital and computational pathology and its current and potential applications in renal cell carcinoma (RCC). EVIDENCE ACQUISITION A systematic review of the English-language literature was conducted using the PubMed, Web of Science, and Scopus databases in December 2022 according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines (PROSPERO ID: CRD42023389282). Risk of bias was assessed according to the Prediction Model Study Risk of Bias Assessment Tool. EVIDENCE SYNTHESIS In total, 20 articles were included in the review. All the studies used a retrospective design, and all digital pathology techniques were implemented retrospectively. The studies were classified according to their primary objective: detection, tumor characterization, and patient outcome. Regarding the transition to clinical practice, several studies showed promising potential. However, none presented a comprehensive assessment of clinical utility and implementation. Notably, there was substantial heterogeneity for both the strategies used for model building and the performance metrics reported. CONCLUSIONS This review highlights the vast potential of digital and computational pathology for the detection, classification, and assessment of oncological outcomes in RCC. Preliminary work in this field has yielded promising results. However, these models have not yet reached a stage where they can be integrated into routine clinical practice. PATIENT SUMMARY Computational pathology combines traditional pathology and technologies such as digital imaging and artificial intelligence to improve diagnosis of disease and identify prognostic factors and new biomarkers. The number of studies exploring its potential in kidney cancer is rapidly increasing. However, despite the surge in research activity, computational pathology is not yet ready for widespread routine use.
Collapse
Affiliation(s)
- Zine-Eddine Khene
- Department of Urology, University of Rennes, Rennes, France; Laboratoire Traitement du Signal et de l'Image, Inserm U1099, Université de Rennes 1, Rennes, France; Department of Urology, UT Southwestern Medical Center, Dallas, TX, USA.
| | - Solène-Florence Kammerer-Jacquet
- Laboratoire Traitement du Signal et de l'Image, Inserm U1099, Université de Rennes 1, Rennes, France; Department of Pathology, University of Rennes, Rennes, France
| | - Pierre Bigot
- Department of Urology, University of Angers, Rennes, France
| | - Noémie Rabilloud
- Laboratoire Traitement du Signal et de l'Image, Inserm U1099, Université de Rennes 1, Rennes, France
| | - Laurence Albiges
- Department of Medical Oncology, Gustave Roussy, Villejuif, France
| | - Vitaly Margulis
- Department of Urology, UT Southwestern Medical Center, Dallas, TX, USA
| | | | - Oscar Acosta
- Laboratoire Traitement du Signal et de l'Image, Inserm U1099, Université de Rennes 1, Rennes, France
| | | | - Yair Lotan
- Department of Urology, UT Southwestern Medical Center, Dallas, TX, USA
| | - Morgan Rouprêt
- Department of Urology, La Pitie Salpétrière Hospital, Paris, France
| | - Karim Bensalah
- Department of Urology, University of Rennes, Rennes, France
| |
Collapse
|
2
|
Ivanova E, Fayzullin A, Grinin V, Ermilov D, Arutyunyan A, Timashev P, Shekhter A. Empowering Renal Cancer Management with AI and Digital Pathology: Pathology, Diagnostics and Prognosis. Biomedicines 2023; 11:2875. [PMID: 38001875 PMCID: PMC10669631 DOI: 10.3390/biomedicines11112875] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/08/2023] [Revised: 09/26/2023] [Accepted: 10/09/2023] [Indexed: 11/26/2023] Open
Abstract
Renal cell carcinoma is a significant health burden worldwide, necessitating accurate and efficient diagnostic methods to guide treatment decisions. Traditional pathology practices have limitations, including interobserver variability and time-consuming evaluations. In recent years, digital pathology tools emerged as a promising solution to enhance the diagnosis and management of renal cancer. This review aims to provide a comprehensive overview of the current state and potential of digital pathology in the context of renal cell carcinoma. Through advanced image analysis algorithms, artificial intelligence (AI) technologies facilitate quantification of cellular and molecular markers, leading to improved accuracy and reproducibility in renal cancer diagnosis. Digital pathology platforms empower remote collaboration between pathologists and help with the creation of comprehensive databases for further research and machine learning applications. The integration of digital pathology tools with other diagnostic modalities, such as radiology and genomics, enables a novel multimodal characterization of different types of renal cell carcinoma. With continuous advancements and refinement, AI technologies are expected to play an integral role in diagnostics and clinical decision-making, improving patient outcomes. In this article, we explored the digital pathology instruments available for clear cell, papillary and chromophobe renal cancers from pathologist and data analyst perspectives.
Collapse
Affiliation(s)
- Elena Ivanova
- Institute for Regenerative Medicine, Sechenov First Moscow State Medical University (Sechenov University), 8-2 Trubetskaya St., Moscow 119991, Russia; (E.I.); (A.F.); (P.T.)
- B. V. Petrovsky Russian Research Center of Surgery, 2 Abrikosovskiy Lane, Moscow 119991, Russia
| | - Alexey Fayzullin
- Institute for Regenerative Medicine, Sechenov First Moscow State Medical University (Sechenov University), 8-2 Trubetskaya St., Moscow 119991, Russia; (E.I.); (A.F.); (P.T.)
| | - Victor Grinin
- PJSC VimpelCom, 10, 8th March Street, Moscow 127083, Russia; (V.G.); (D.E.); (A.A.)
| | - Dmitry Ermilov
- PJSC VimpelCom, 10, 8th March Street, Moscow 127083, Russia; (V.G.); (D.E.); (A.A.)
| | - Alexander Arutyunyan
- PJSC VimpelCom, 10, 8th March Street, Moscow 127083, Russia; (V.G.); (D.E.); (A.A.)
| | - Peter Timashev
- Institute for Regenerative Medicine, Sechenov First Moscow State Medical University (Sechenov University), 8-2 Trubetskaya St., Moscow 119991, Russia; (E.I.); (A.F.); (P.T.)
| | - Anatoly Shekhter
- Institute for Regenerative Medicine, Sechenov First Moscow State Medical University (Sechenov University), 8-2 Trubetskaya St., Moscow 119991, Russia; (E.I.); (A.F.); (P.T.)
| |
Collapse
|
3
|
Dent A, Faust K, Lam K, Alhangari N, Leon AJ, Tsang Q, Kamil ZS, Gao A, Pal P, Lheureux S, Oza A, Diamandis P. HAVOC: Small-scale histomic mapping of cancer biodiversity across large tissue distances using deep neural networks. SCIENCE ADVANCES 2023; 9:eadg1894. [PMID: 37774029 PMCID: PMC10541015 DOI: 10.1126/sciadv.adg1894] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 12/07/2022] [Accepted: 08/28/2023] [Indexed: 10/01/2023]
Abstract
Intratumoral heterogeneity can wreak havoc on current precision medicine strategies because of challenges in sufficient sampling of geographically separated areas of biodiversity distributed across centimeter-scale tumor distances. To address this gap, we developed a deep learning pipeline that leverages histomorphologic fingerprints of tissue to create "Histomic Atlases of Variation Of Cancers" (HAVOC). Using a number of objective molecular readouts, we demonstrate that HAVOC can define regional cancer boundaries with distinct biology. Using larger tumor specimens, we show that HAVOC can map biodiversity even across multiple tissue sections. By guiding profiling of 19 partitions across six high-grade gliomas, HAVOC revealed that distinct differentiation states can often coexist and be regionally distributed within these tumors. Last, to highlight generalizability, we benchmark HAVOC on additional tumor types. Together, we establish HAVOC as a versatile tool to generate small-scale maps of tissue heterogeneity and guide regional deployment of molecular resources to relevant biodiverse niches.
Collapse
Affiliation(s)
- Anglin Dent
- Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, ON M5S 1A8, Canada
| | - Kevin Faust
- Department of Computer Science, University of Toronto, 40 St. George Street, Toronto, ON M5S 2E4, Canada
- Princess Margaret Cancer Centre, 101 College Street, Toronto, ON M5G 1L7, Canada
| | - K. H. Brian Lam
- Princess Margaret Cancer Centre, 101 College Street, Toronto, ON M5G 1L7, Canada
- Department of Pathology and Laboratory Medicine, David Geffen School of Medicine, University of California Los Angeles, Los Angeles, CA, USA
| | - Narges Alhangari
- Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, ON M5S 1A8, Canada
| | - Alberto J. Leon
- Princess Margaret Cancer Centre, 101 College Street, Toronto, ON M5G 1L7, Canada
| | - Queenie Tsang
- Princess Margaret Cancer Centre, 101 College Street, Toronto, ON M5G 1L7, Canada
| | - Zaid Saeed Kamil
- Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, ON M5S 1A8, Canada
- Laboratory Medicine Program, Department of Pathology, University Health Network, 200 Elizabeth Street, Toronto, ON M5G 2C4, Canada
| | - Andrew Gao
- Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, ON M5S 1A8, Canada
- Laboratory Medicine Program, Department of Pathology, University Health Network, 200 Elizabeth Street, Toronto, ON M5G 2C4, Canada
| | - Prodipto Pal
- Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, ON M5S 1A8, Canada
- Laboratory Medicine Program, Department of Pathology, University Health Network, 200 Elizabeth Street, Toronto, ON M5G 2C4, Canada
| | - Stephanie Lheureux
- Princess Margaret Cancer Centre, 101 College Street, Toronto, ON M5G 1L7, Canada
| | - Amit Oza
- Princess Margaret Cancer Centre, 101 College Street, Toronto, ON M5G 1L7, Canada
| | - Phedias Diamandis
- Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, ON M5S 1A8, Canada
- Princess Margaret Cancer Centre, 101 College Street, Toronto, ON M5G 1L7, Canada
- Laboratory Medicine Program, Department of Pathology, University Health Network, 200 Elizabeth Street, Toronto, ON M5G 2C4, Canada
- Department of Medical Biophysics, University of Toronto, 101 College St, Toronto, ON M5G 1L7, Canada
| |
Collapse
|
4
|
Paliwal A, Faust K, Alshoumer A, Diamandis P. Standardizing analysis of intra-tumoral heterogeneity with computational pathology. Genes Chromosomes Cancer 2023; 62:526-539. [PMID: 37067005 DOI: 10.1002/gcc.23146] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/15/2023] [Revised: 03/28/2023] [Accepted: 04/04/2023] [Indexed: 04/18/2023] Open
Abstract
Many malignant cancers like glioblastoma are highly adaptive diseases that dynamically change their regional biology to survive and thrive under diverse microenvironmental and therapeutic pressures. While the concept of intra-tumoral heterogeneity has become a major paradigm in cancer research and care, systematic approaches to assess and document bio-variation in cancer are still in their infancy. Here we discuss existing approaches and challenges to documenting intra-tumoral heterogeneity and emerging computational approaches that leverage artificial intelligence to begin to overcome these limitations. We propose how these emerging techniques can be coupled with a diversity of molecular tools to address intra-tumoral heterogeneity more systematically in research and in practice, especially across larger specimens and longitudinal analyses. Systematic documentation and characterization of heterogeneity across entire tumor specimens and their longitudinal evolution has the potential to improve our understanding and treatment of cancer.
Collapse
Affiliation(s)
- Ameesha Paliwal
- Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, Ontario, Canada
| | - Kevin Faust
- Department of Computer Science, University of Toronto, Toronto, Ontario, Canada
- Department of Laboratory Medicine and Pathology, Princess Margaret Cancer Centre, Toronto, Ontario, Canada
| | - Azhar Alshoumer
- Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, Ontario, Canada
- Laboratory Medicine Program, Department of Pathology, University Health Network, Toronto, Ontario, Canada
| | - Phedias Diamandis
- Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, Ontario, Canada
- Department of Laboratory Medicine and Pathology, Princess Margaret Cancer Centre, Toronto, Ontario, Canada
- Laboratory Medicine Program, Department of Pathology, University Health Network, Toronto, Ontario, Canada
- Department of Medical Biophysics, University of Toronto, Toronto, Ontario, Canada
| |
Collapse
|
5
|
Distante A, Marandino L, Bertolo R, Ingels A, Pavan N, Pecoraro A, Marchioni M, Carbonara U, Erdem S, Amparore D, Campi R, Roussel E, Caliò A, Wu Z, Palumbo C, Borregales LD, Mulders P, Muselaers CHJ. Artificial Intelligence in Renal Cell Carcinoma Histopathology: Current Applications and Future Perspectives. Diagnostics (Basel) 2023; 13:2294. [PMID: 37443687 DOI: 10.3390/diagnostics13132294] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/31/2023] [Revised: 07/01/2023] [Accepted: 07/04/2023] [Indexed: 07/15/2023] Open
Abstract
Renal cell carcinoma (RCC) is characterized by its diverse histopathological features, which pose possible challenges to accurate diagnosis and prognosis. A comprehensive literature review was conducted to explore recent advancements in the field of artificial intelligence (AI) in RCC pathology. The aim of this paper is to assess whether these advancements hold promise in improving the precision, efficiency, and objectivity of histopathological analysis for RCC, while also reducing costs and interobserver variability and potentially alleviating the labor and time burden experienced by pathologists. The reviewed AI-powered approaches demonstrate effective identification and classification abilities regarding several histopathological features associated with RCC, facilitating accurate diagnosis, grading, and prognosis prediction and enabling precise and reliable assessments. Nevertheless, implementing AI in renal cell carcinoma generates challenges concerning standardization, generalizability, benchmarking performance, and integration of data into clinical workflows. Developing methodologies that enable pathologists to interpret AI decisions accurately is imperative. Moreover, establishing more robust and standardized validation workflows is crucial to instill confidence in AI-powered systems' outcomes. These efforts are vital for advancing current state-of-the-art practices and enhancing patient care in the future.
Collapse
Affiliation(s)
- Alfredo Distante
- Department of Urology, Catholic University of the Sacred Heart, 00168 Roma, Italy
- Department of Urology, Radboud University Medical Center, Geert Grooteplein 10, 6525 GA Nijmegen, The Netherlands
| | - Laura Marandino
- Department of Medical Oncology, IRCCS Ospedale San Raffaele, 20132 Milan, Italy
| | - Riccardo Bertolo
- Department of Urology, San Carlo Di Nancy Hospital, 00165 Rome, Italy
| | - Alexandre Ingels
- Department of Urology, University Hospital Henri Mondor, APHP (Assistance Publique-Hôpitaux de Paris), 94000 Créteil, France
| | - Nicola Pavan
- Department of Surgical, Oncological and Oral Sciences, Section of Urology, University of Palermo, 90133 Palermo, Italy
| | - Angela Pecoraro
- Department of Urology, San Luigi Gonzaga Hospital, University of Turin, Orbassano, 10043 Turin, Italy
| | - Michele Marchioni
- Department of Medical, Oral and Biotechnological Sciences, G. d'Annunzio University of Chieti, 66100 Chieti, Italy
| | - Umberto Carbonara
- Andrology and Kidney Transplantation Unit, Department of Emergency and Organ Transplantation-Urology, University of Bari, 70121 Bari, Italy
| | - Selcuk Erdem
- Division of Urologic Oncology, Department of Urology, Istanbul University Istanbul Faculty of Medicine, Istanbul 34093, Turkey
| | - Daniele Amparore
- Department of Urology, San Luigi Gonzaga Hospital, University of Turin, Orbassano, 10043 Turin, Italy
| | - Riccardo Campi
- Urological Robotic Surgery and Renal Transplantation Unit, Careggi Hospital, University of Florence, 50121 Firenze, Italy
| | - Eduard Roussel
- Department of Urology, University Hospitals Leuven, 3000 Leuven, Belgium
| | - Anna Caliò
- Section of Pathology, Department of Diagnostic and Public Health, University of Verona, 37134 Verona, Italy
| | - Zhenjie Wu
- Department of Urology, Changhai Hospital, Naval Medical University, Shanghai 200433, China
| | - Carlotta Palumbo
- Division of Urology, Maggiore della Carità Hospital of Novara, Department of Translational Medicine, University of Eastern Piedmont, 13100 Novara, Italy
| | - Leonardo D Borregales
- Department of Urology, Well Cornell Medicine, New York-Presbyterian Hospital, New York, NY 10032, USA
| | - Peter Mulders
- Department of Urology, Radboud University Medical Center, Geert Grooteplein 10, 6525 GA Nijmegen, The Netherlands
| | - Constantijn H J Muselaers
- Department of Urology, Radboud University Medical Center, Geert Grooteplein 10, 6525 GA Nijmegen, The Netherlands
| |
Collapse
|
6
|
Zhang H, Botler M, Kooman JP. Deep Learning for Image Analysis in Kidney Care. ADVANCES IN KIDNEY DISEASE AND HEALTH 2023; 30:25-32. [PMID: 36723278 DOI: 10.1053/j.akdh.2022.11.003] [Citation(s) in RCA: 1] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Subscribe] [Scholar Register] [Received: 04/15/2022] [Revised: 08/23/2022] [Accepted: 11/07/2022] [Indexed: 12/24/2022]
Abstract
Analysis of medical images, such as radiological or tissue specimens, is an indispensable part of medical diagnostics. Conventionally done manually, the process may sometimes be time-consuming and prone to interobserver variability. Image classification and segmentation by deep learning strategies, predominantly convolutional neural networks, may provide a significant advance in the diagnostic process. In renal medicine, most evidence has been generated around the radiological assessment of renal abnormalities and histological analysis of renal biopsy specimens' segmentation. In this article, the basic principles of image analysis by convolutional neural networks, brief descriptions of convolutional neural networks, and their system architecture for image analysis are discussed, in combination with examples regarding their use in image analysis in nephrology.
Collapse
Affiliation(s)
| | | | - Jeroen P Kooman
- Division of Nephrology, Department of Internal Medicine, Maastricht University Medical Center, Maastricht, The Netherlands
| |
Collapse
|
7
|
Wessels F, Kuntz S, Krieghoff-Henning E, Schmitt M, Braun V, Worst TS, Neuberger M, Steeg M, Gaiser T, Fröhling S, Michel MS, Nuhn P, Brinker TJ. Artificial intelligence to predict oncological outcome directly from hematoxylin and eosin-stained slides in urology. Minerva Urol Nephrol 2022; 74:538-550. [PMID: 35274903 DOI: 10.23736/s2724-6051.22.04758-9] [Citation(s) in RCA: 8] [Impact Index Per Article: 4.0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/08/2022]
Abstract
INTRODUCTION Artificial intelligence (AI) has been successfully applied for automatic tumor detection and grading in histopathological image analysis in urologic oncology. The aim of this review was to assess the applicability of these approaches in image-based oncological outcome prediction. EVIDENCE ACQUISITION A systematic literature search was conducted using the databases MEDLINE through PubMed and Web of Science up to April 20, 2021. Studies investigating AI approaches to determine the risk of recurrence, metastasis, or survival directly from H&E-stained tissue sections in prostate, renal cell or urothelial carcinoma were included. Characteristics of the AI approach and performance metrics were extracted and summarized. Risk of bias (RoB) was assessed using the PROBAST tool. EVIDENCE SYNTHESIS 16 studies yielding a total of 6658 patients and reporting on 17 outcome predictions were included. Six studies focused on renal cell, six on prostate and three on urothelial carcinoma while one study investigated renal cell and urothelial carcinoma. Handcrafted feature extraction was used in five, a convolutional neural network (CNN) in six and a deep feature extraction in four studies. One study compared a CNN with handcrafted feature extraction. In seven outcome predictions, a multivariable comparison with clinicopathological parameters was reported. Five of them showed statistically significant hazard ratios for the AI's model's-prediction. However, RoB was high in 15 outcome predictions and unclear in two. CONCLUSIONS The included studies are promising but predominantly early pilot studies, therefore primarily highlighting the potential of AI approaches. Additional well-designed studies are needed to assess the actual clinical applicability.
Collapse
Affiliation(s)
- Frederik Wessels
- Digital Biomarkers for Oncology Group, National Center for Tumor Diseases (NCT), German Cancer Research Center (DKFZ), Heidelberg, Germany.,Department of Urology and Urological Surgery, Medical Faculty Mannheim of Heidelberg University, University Medical Center Mannheim, Mannheim, Germany
| | - Sara Kuntz
- Digital Biomarkers for Oncology Group, National Center for Tumor Diseases (NCT), German Cancer Research Center (DKFZ), Heidelberg, Germany
| | - Eva Krieghoff-Henning
- Digital Biomarkers for Oncology Group, National Center for Tumor Diseases (NCT), German Cancer Research Center (DKFZ), Heidelberg, Germany
| | - Max Schmitt
- Digital Biomarkers for Oncology Group, National Center for Tumor Diseases (NCT), German Cancer Research Center (DKFZ), Heidelberg, Germany
| | - Volker Braun
- Library for the Medical Faculty Mannheim of Heidelberg University, Mannheim, Germany
| | - Thomas S Worst
- Department of Urology and Urological Surgery, Medical Faculty Mannheim of Heidelberg University, University Medical Center Mannheim, Mannheim, Germany
| | - Manuel Neuberger
- Department of Urology and Urological Surgery, Medical Faculty Mannheim of Heidelberg University, University Medical Center Mannheim, Mannheim, Germany
| | - Matthias Steeg
- Institute of Pathology, Medical Faculty Mannheim of Heidelberg University, University Medical Center Mannheim, Mannheim, Germany
| | - Timo Gaiser
- Institute of Pathology, Medical Faculty Mannheim of Heidelberg University, University Medical Center Mannheim, Mannheim, Germany
| | - Stefan Fröhling
- National Center for Tumor Diseases, German Cancer Research Center, Heidelberg, Germany
| | - Maurice-Stephan Michel
- Department of Urology and Urological Surgery, Medical Faculty Mannheim of Heidelberg University, University Medical Center Mannheim, Mannheim, Germany
| | - Philipp Nuhn
- Department of Urology and Urological Surgery, Medical Faculty Mannheim of Heidelberg University, University Medical Center Mannheim, Mannheim, Germany
| | - Titus J Brinker
- Digital Biomarkers for Oncology Group, National Center for Tumor Diseases (NCT), German Cancer Research Center (DKFZ), Heidelberg, Germany -
| |
Collapse
|
8
|
Deep learning can predict survival directly from histology in clear cell renal cell carcinoma. PLoS One 2022; 17:e0272656. [PMID: 35976907 PMCID: PMC9385058 DOI: 10.1371/journal.pone.0272656] [Citation(s) in RCA: 8] [Impact Index Per Article: 4.0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/09/2022] [Accepted: 07/24/2022] [Indexed: 11/19/2022] Open
Abstract
For clear cell renal cell carcinoma (ccRCC) risk-dependent diagnostic and therapeutic algorithms are routinely implemented in clinical practice. Artificial intelligence-based image analysis has the potential to improve outcome prediction and thereby risk stratification. Thus, we investigated whether a convolutional neural network (CNN) can extract relevant image features from a representative hematoxylin and eosin-stained slide to predict 5-year overall survival (5y-OS) in ccRCC. The CNN was trained to predict 5y-OS in a binary manner using slides from TCGA and validated using an independent in-house cohort. Multivariable logistic regression was used to combine of the CNNs prediction and clinicopathological parameters. A mean balanced accuracy of 72.0% (standard deviation [SD] = 7.9%), sensitivity of 72.4% (SD = 10.6%), specificity of 71.7% (SD = 11.9%) and area under receiver operating characteristics curve (AUROC) of 0.75 (SD = 0.07) was achieved on the TCGA training set (n = 254 patients / WSIs) using 10-fold cross-validation. On the external validation cohort (n = 99 patients / WSIs), mean accuracy, sensitivity, specificity and AUROC were 65.5% (95%-confidence interval [CI]: 62.9–68.1%), 86.2% (95%-CI: 81.8–90.5%), 44.9% (95%-CI: 40.2–49.6%), and 0.70 (95%-CI: 0.69–0.71). A multivariable model including age, tumor stage and metastasis yielded an AUROC of 0.75 on the TCGA cohort. The inclusion of the CNN-based classification (Odds ratio = 4.86, 95%-CI: 2.70–8.75, p < 0.01) raised the AUROC to 0.81. On the validation cohort, both models showed an AUROC of 0.88. In univariable Cox regression, the CNN showed a hazard ratio of 3.69 (95%-CI: 2.60–5.23, p < 0.01) on TCGA and 2.13 (95%-CI: 0.92–4.94, p = 0.08) on external validation. The results demonstrate that the CNN’s image-based prediction of survival is promising and thus this widely applicable technique should be further investigated with the aim of improving existing risk stratification in ccRCC.
Collapse
|
9
|
Acosta PH, Panwar V, Jarmale V, Christie A, Jasti J, Margulis V, Rakheja D, Cheville J, Leibovich BC, Parker A, Brugarolas J, Kapur P, Rajaram S. Intratumoral Resolution of Driver Gene Mutation Heterogeneity in Renal Cancer Using Deep Learning. Cancer Res 2022; 82:2792-2806. [PMID: 35654752 PMCID: PMC9373732 DOI: 10.1158/0008-5472.can-21-2318] [Citation(s) in RCA: 7] [Impact Index Per Article: 3.5] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/15/2021] [Revised: 11/16/2021] [Accepted: 05/27/2022] [Indexed: 02/05/2023]
Abstract
Intratumoral heterogeneity arising from tumor evolution poses significant challenges biologically and clinically. Dissecting this complexity may benefit from deep learning (DL) algorithms, which can infer molecular features from ubiquitous hematoxylin and eosin (H&E)-stained tissue sections. Although DL algorithms have been developed to predict some driver mutations from H&E images, the ability of these DL algorithms to resolve intratumoral mutation heterogeneity at subclonal spatial resolution is unexplored. Here, we apply DL to a paradigm of intratumoral heterogeneity, clear cell renal cell carcinoma (ccRCC), the most common type of kidney cancer. Matched IHC and H&E images were leveraged to develop DL models for predicting intratumoral genetic heterogeneity of the three most frequently mutated ccRCC genes, BAP1, PBRM1, and SETD2. DL models were generated on a large cohort (N = 1,282) and tested on several independent cohorts, including a TCGA cohort (N = 363 patients) and two tissue microarray (TMA) cohorts (N = 118 and 365 patients). These models were also expanded to a patient-derived xenograft (PDX) TMA, affording analysis of homotopic and heterotopic interactions of tumor and stroma. The status of all three genes could be inferred by DL, with BAP1 showing the highest sensitivity and performance within and across tissue samples (AUC = 0.87-0.89 on holdout). BAP1 results were validated on independent human (AUC = 0.77-0.84) and PDX (AUC = 0.80) cohorts. Finally, BAP1 predictions correlated with clinical outputs such as disease-specific survival. Overall, these data show that DL models can resolve intratumoral heterogeneity in cancer with potential diagnostic, prognostic, and biological implications. SIGNIFICANCE This work demonstrates the potential for deep learning analysis of histopathologic images to serve as a fast, low-cost method to assess genetic intratumoral heterogeneity. See related commentary by Song et al., p. 2672.
Collapse
Affiliation(s)
- Paul H. Acosta
- Lyda Hill Department of Bioinformatics, University of Texas Southwestern Medical Center, Dallas, TX, USA
| | - Vandana Panwar
- Department of Pathology, University of Texas Southwestern Medical Center, Dallas, TX, USA
| | - Vipul Jarmale
- Lyda Hill Department of Bioinformatics, University of Texas Southwestern Medical Center, Dallas, TX, USA
| | - Alana Christie
- Kidney Cancer Program, Simmons Comprehensive Cancer Center, University of Texas Southwestern Medical Center, Dallas, TX, USA
| | - Jay Jasti
- Lyda Hill Department of Bioinformatics, University of Texas Southwestern Medical Center, Dallas, TX, USA
| | - Vitaly Margulis
- Kidney Cancer Program, Simmons Comprehensive Cancer Center, University of Texas Southwestern Medical Center, Dallas, TX, USA,Department of Urology, University of Texas Southwestern Medical Center at Dallas, Dallas, TX, USA
| | - Dinesh Rakheja
- Department of Pathology, University of Texas Southwestern Medical Center, Dallas, TX, USA
| | - John Cheville
- Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, MN, USA
| | | | - Alexander Parker
- Office of Research Affairs, University of Florida, College of Medicine, Jacksonville, FL, USA
| | - James Brugarolas
- Kidney Cancer Program, Simmons Comprehensive Cancer Center, University of Texas Southwestern Medical Center, Dallas, TX, USA,Department of Internal Medicine (Hematology-Oncology), University of Texas Southwestern Medical Center, Dallas, TX, USA
| | - Payal Kapur
- Department of Pathology, University of Texas Southwestern Medical Center, Dallas, TX, USA,Kidney Cancer Program, Simmons Comprehensive Cancer Center, University of Texas Southwestern Medical Center, Dallas, TX, USA,Department of Urology, University of Texas Southwestern Medical Center at Dallas, Dallas, TX, USA.,Corresponding Authors: 1) Payal Kapur, MD, Department of Pathology, University of Texas Southwestern Medical Center, 5323 Harry Hines Boulevard, Dallas, TX 75390, Ph: 214.590.6592, 2) Satwik Rajaram, PhD, Lyda Hill Department of Bioinformatics, University of Texas Southwestern Medical Center, 5323 Harry Hines Boulevard, Dallas, TX - 75390-9365, Ph: 214-645-1727,
| | - Satwik Rajaram
- Lyda Hill Department of Bioinformatics, University of Texas Southwestern Medical Center, Dallas, TX, USA,Department of Pathology, University of Texas Southwestern Medical Center, Dallas, TX, USA,Kidney Cancer Program, Simmons Comprehensive Cancer Center, University of Texas Southwestern Medical Center, Dallas, TX, USA,Corresponding Authors: 1) Payal Kapur, MD, Department of Pathology, University of Texas Southwestern Medical Center, 5323 Harry Hines Boulevard, Dallas, TX 75390, Ph: 214.590.6592, 2) Satwik Rajaram, PhD, Lyda Hill Department of Bioinformatics, University of Texas Southwestern Medical Center, 5323 Harry Hines Boulevard, Dallas, TX - 75390-9365, Ph: 214-645-1727,
| |
Collapse
|
10
|
Park J, Barahona‐Torres N, Jang S, Mok KY, Kim HJ, Han S, Cho K, Zhou X, Fu AKY, Ip NY, Seo J, Choi M, Jeong H, Hwang D, Lee DY, Byun MS, Yi D, Han JW, Mook‐Jung I, Hardy J. Multi-Omics-Based Autophagy-Related Untypical Subtypes in Patients with Cerebral Amyloid Pathology. ADVANCED SCIENCE (WEINHEIM, BADEN-WURTTEMBERG, GERMANY) 2022; 9:e2201212. [PMID: 35694866 PMCID: PMC9376815 DOI: 10.1002/advs.202201212] [Citation(s) in RCA: 6] [Impact Index Per Article: 3.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 03/02/2022] [Revised: 04/26/2022] [Indexed: 05/05/2023]
Abstract
Recent multi-omics analyses paved the way for a comprehensive understanding of pathological processes. However, only few studies have explored Alzheimer's disease (AD) despite the possibility of biological subtypes within these patients. For this study, unsupervised classification of four datasets (genetics, miRNA transcriptomics, proteomics, and blood-based biomarkers) using Multi-Omics Factor Analysis+ (MOFA+), along with systems-biological approaches following various downstream analyses are performed. New subgroups within 170 patients with cerebral amyloid pathology (Aβ+) are revealed and the features of them are identified based on the top-rated targets constructing multi-omics factors of both whole (M-TPAD) and immune-focused models (M-IPAD). The authors explored the characteristics of subtypes and possible key-drivers for AD pathogenesis. Further in-depth studies showed that these subtypes are associated with longitudinal brain changes and autophagy pathways are main contributors. The significance of autophagy or clustering tendency is validated in peripheral blood mononuclear cells (PBMCs; n = 120 including 30 Aβ- and 90 Aβ+), induced pluripotent stem cell-derived human brain organoids/microglia (n = 12 including 5 Aβ-, 5 Aβ+, and CRISPR-Cas9 apolipoprotein isogenic lines), and human brain transcriptome (n = 78). Collectively, this study provides a strategy for precision medicine therapy and drug development for AD using integrative multi-omics analysis and network modelling.
Collapse
Affiliation(s)
- Jong‐Chan Park
- Department of Neurodegenerative DiseaseUCL Queen Square Institute of NeurologyUniversity College LondonLondonWC1N 3BGUK
- Department of Biochemistry and Biomedical SciencesCollege of MedicineSeoul National UniversitySeoul03080Republic of Korea
- Neuroscience Research InstituteMedical Research CenterCollege of MedicineSeoul National UniversitySeoul03080Republic of Korea
- SNU Korea Dementia Research CenterCollege of MedicineSeoul National UniversitySeoul03080Republic of Korea
| | - Natalia Barahona‐Torres
- Department of Neurodegenerative DiseaseUCL Queen Square Institute of NeurologyUniversity College LondonLondonWC1N 3BGUK
| | - So‐Yeong Jang
- Department of Bio and Brain EngineeringKorea Advanced Institute of Science and TechnologyDaejeon34141Republic of Korea
| | - Kin Y. Mok
- Department of Neurodegenerative DiseaseUCL Queen Square Institute of NeurologyUniversity College LondonLondonWC1N 3BGUK
| | - Haeng Jun Kim
- Department of Biochemistry and Biomedical SciencesCollege of MedicineSeoul National UniversitySeoul03080Republic of Korea
- SNU Korea Dementia Research CenterCollege of MedicineSeoul National UniversitySeoul03080Republic of Korea
| | - Sun‐Ho Han
- Department of Biochemistry and Biomedical SciencesCollege of MedicineSeoul National UniversitySeoul03080Republic of Korea
- Neuroscience Research InstituteMedical Research CenterCollege of MedicineSeoul National UniversitySeoul03080Republic of Korea
- SNU Korea Dementia Research CenterCollege of MedicineSeoul National UniversitySeoul03080Republic of Korea
| | - Kwang‐Hyun Cho
- Department of Bio and Brain EngineeringKorea Advanced Institute of Science and TechnologyDaejeon34141Republic of Korea
| | - Xiaopu Zhou
- Division of Life ScienceState Key Laboratory of Molecular NeuroscienceMolecular Neuroscience CenterThe Hong Kong University of Science and TechnologyClear Water Bay, KowloonHong Kong999077China
- Hong Kong Center for Neurodegenerative DiseasesHong Kong Science ParkHong Kong999077China
- Guangdong Provincial Key Laboratory of Brain ScienceDisease and Drug DevelopmentHKUST Shenzhen Research InstituteShenzhen‐Hong Kong Institute of Brain ScienceShenzhenGuangdong518057China
| | - Amy K. Y. Fu
- Division of Life ScienceState Key Laboratory of Molecular NeuroscienceMolecular Neuroscience CenterThe Hong Kong University of Science and TechnologyClear Water Bay, KowloonHong Kong999077China
- Hong Kong Center for Neurodegenerative DiseasesHong Kong Science ParkHong Kong999077China
- Guangdong Provincial Key Laboratory of Brain ScienceDisease and Drug DevelopmentHKUST Shenzhen Research InstituteShenzhen‐Hong Kong Institute of Brain ScienceShenzhenGuangdong518057China
| | - Nancy Y. Ip
- Division of Life ScienceState Key Laboratory of Molecular NeuroscienceMolecular Neuroscience CenterThe Hong Kong University of Science and TechnologyClear Water Bay, KowloonHong Kong999077China
- Hong Kong Center for Neurodegenerative DiseasesHong Kong Science ParkHong Kong999077China
- Guangdong Provincial Key Laboratory of Brain ScienceDisease and Drug DevelopmentHKUST Shenzhen Research InstituteShenzhen‐Hong Kong Institute of Brain ScienceShenzhenGuangdong518057China
| | - Jieun Seo
- Department of Laboratory MedicineSeverance HospitalYonsei University College of MedicineSeoul03722Republic of Korea
| | - Murim Choi
- Department of Biochemistry and Biomedical SciencesCollege of MedicineSeoul National UniversitySeoul03080Republic of Korea
| | - Hyobin Jeong
- European Molecular Biology LaboratoryGenome Biology UnitHeidelberg69117Germany
| | - Daehee Hwang
- Department of Biological SciencesSeoul National UniversitySeoul08826Republic of Korea
| | - Dong Young Lee
- Institute of Human Behavioral MedicineMedical Research CenterSeoul National UniversitySeoul03080Republic of Korea
- Department of PsychiatryCollege of medicineSeoul National UniversitySeoul03080Republic of Korea
- Department of NeuropsychiatrySeoul National University HospitalSeoul03080Republic of Korea
| | - Min Soo Byun
- Department of PsychiatryPusan National University Yangsan HospitalYangsan50612Republic of Korea
| | - Dahyun Yi
- Biomedical Research InstituteSeoul National University HospitalSeoul03082Republic of Korea
| | - Jong Won Han
- Department of Biochemistry and Biomedical SciencesCollege of MedicineSeoul National UniversitySeoul03080Republic of Korea
| | - Inhee Mook‐Jung
- Department of Biochemistry and Biomedical SciencesCollege of MedicineSeoul National UniversitySeoul03080Republic of Korea
- Neuroscience Research InstituteMedical Research CenterCollege of MedicineSeoul National UniversitySeoul03080Republic of Korea
- SNU Korea Dementia Research CenterCollege of MedicineSeoul National UniversitySeoul03080Republic of Korea
| | - John Hardy
- Department of Neurodegenerative DiseaseUCL Queen Square Institute of NeurologyUniversity College LondonLondonWC1N 3BGUK
| |
Collapse
|
11
|
Foroughi pour A, White BS, Park J, Sheridan TB, Chuang JH. Deep learning features encode interpretable morphologies within histological images. Sci Rep 2022; 12:9428. [PMID: 35676395 PMCID: PMC9177767 DOI: 10.1038/s41598-022-13541-2] [Citation(s) in RCA: 4] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/25/2022] [Accepted: 05/25/2022] [Indexed: 12/12/2022] Open
Abstract
Convolutional neural networks (CNNs) are revolutionizing digital pathology by enabling machine learning-based classification of a variety of phenotypes from hematoxylin and eosin (H&E) whole slide images (WSIs), but the interpretation of CNNs remains difficult. Most studies have considered interpretability in a post hoc fashion, e.g. by presenting example regions with strongly predicted class labels. However, such an approach does not explain the biological features that contribute to correct predictions. To address this problem, here we investigate the interpretability of H&E-derived CNN features (the feature weights in the final layer of a transfer-learning-based architecture). While many studies have incorporated CNN features into predictive models, there has been little empirical study of their properties. We show such features can be construed as abstract morphological genes (“mones”) with strong independent associations to biological phenotypes. Many mones are specific to individual cancer types, while others are found in multiple cancers especially from related tissue types. We also observe that mone-mone correlations are strong and robustly preserved across related cancers. Importantly, linear mone-based classifiers can very accurately separate 38 distinct classes (19 tumor types and their adjacent normals, AUC = \documentclass[12pt]{minimal}
\usepackage{amsmath}
\usepackage{wasysym}
\usepackage{amsfonts}
\usepackage{amssymb}
\usepackage{amsbsy}
\usepackage{mathrsfs}
\usepackage{upgreek}
\setlength{\oddsidemargin}{-69pt}
\begin{document}$$97.1\% \pm 2.8\%$$\end{document}97.1%±2.8% for each class prediction), and linear classifiers are also highly effective for universal tumor detection (AUC = \documentclass[12pt]{minimal}
\usepackage{amsmath}
\usepackage{wasysym}
\usepackage{amsfonts}
\usepackage{amssymb}
\usepackage{amsbsy}
\usepackage{mathrsfs}
\usepackage{upgreek}
\setlength{\oddsidemargin}{-69pt}
\begin{document}$$99.2\% \pm 0.12\%$$\end{document}99.2%±0.12%). This linearity provides evidence that individual mones or correlated mone clusters may be associated with interpretable histopathological features or other patient characteristics. In particular, the statistical similarity of mones to gene expression values allows integrative mone analysis via expression-based bioinformatics approaches. We observe strong correlations between individual mones and individual gene expression values, notably mones associated with collagen gene expression in ovarian cancer. Mone-expression comparisons also indicate that immunoglobulin expression can be identified using mones in colon adenocarcinoma and that immune activity can be identified across multiple cancer types, and we verify these findings by expert histopathological review. Our work demonstrates that mones provide a morphological H&E decomposition that can be effectively associated with diverse phenotypes, analogous to the interpretability of transcription via gene expression values. Our work also demonstrates mones can be interpreted without using a classifier as a proxy.
Collapse
|
12
|
Dent A, Diamandis P. Integrating computational pathology and proteomics to address tumor heterogeneity. J Pathol 2022; 257:445-453. [DOI: 10.1002/path.5905] [Citation(s) in RCA: 2] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/11/2022] [Revised: 03/20/2022] [Accepted: 03/30/2022] [Indexed: 11/11/2022]
Affiliation(s)
- Anglin Dent
- Department of Laboratory Medicine and Pathobiology University of Toronto Toronto Ontario M5S 1A8 Canada
- Princess Margaret Cancer Center University Health Network, Toronto, Ontario, 610 University Avenue, M5G 2C1 Canada
| | - Phedias Diamandis
- Department of Laboratory Medicine and Pathobiology University of Toronto Toronto Ontario M5S 1A8 Canada
- Princess Margaret Cancer Center University Health Network, Toronto, Ontario, 610 University Avenue, M5G 2C1 Canada
- Laboratory Medicine Program University Health Network, 200 Elizabeth Street, Toronto, ON Toronto Ontario M5G 2C4 Canada
- Department of Medical Biophysics University of Toronto Toronto Ontario M5S 1A8 Canada
| |
Collapse
|
13
|
Faust K, Lee MK, Dent A, Fiala C, Portante A, Rabindranath M, Alsafwani N, Gao A, Djuric U, Diamandis P. Integrating morphologic and molecular histopathological features through whole slide image registration and deep learning. Neurooncol Adv 2022; 4:vdac001. [PMID: 35156037 PMCID: PMC8826810 DOI: 10.1093/noajnl/vdac001] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/14/2022] Open
Abstract
Abstract
Background
Modern molecular pathology workflows in neuro-oncology heavily rely on the integration of morphologic and immunohistochemical patterns for analysis, classification, and prognostication. However, despite the recent emergence of digital pathology platforms and artificial intelligence-driven computational image analysis tools, automating the integration of histomorphologic information found across these multiple studies is challenged by large files sizes of whole slide images (WSIs) and shifts/rotations in tissue sections introduced during slide preparation.
Methods
To address this, we develop a workflow that couples different computer vision tools including scale-invariant feature transform (SIFT) and deep learning to efficiently align and integrate histopathological information found across multiple independent studies. We highlight the utility and automation potential of this workflow in the molecular subclassification and discovery of previously unappreciated spatial patterns in diffuse gliomas.
Results
First, we show how a SIFT-driven computer vision workflow was effective at automated WSI alignment in a cohort of 107 randomly selected surgical neuropathology cases (97/107 (91%) showing appropriate matches, AUC = 0.96). This alignment allows our AI-driven diagnostic workflow to not only differentiate different brain tumor types, but also integrate and carry out molecular subclassification of diffuse gliomas using relevant immunohistochemical biomarkers (IDH1-R132H, ATRX). To highlight the discovery potential of this workflow, we also examined spatial distributions of tumors showing heterogenous expression of the proliferation marker MIB1 and Olig2. This analysis helped uncovered an interesting and unappreciated association of Olig2 positive and proliferative areas in some gliomas (r = 0.62).
Conclusion
This efficient neuropathologist-inspired workflow provides a generalizable approach to help automate a variety of advanced immunohistochemically compatible diagnostic and discovery exercises in surgical neuropathology and neuro-oncology.
Collapse
Affiliation(s)
- Kevin Faust
- Department of Computer Science, University of Toronto, 40 St. George Street, Toronto, ON M5S 2E4, Canada
- Laboratory Medicine Program, Department of Pathology, University Health Network, 200 Elizabeth Street, Toronto, ON M5G 2C4, Canada
| | - Michael K Lee
- Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, ON M5S 1A8, Canada
| | - Anglin Dent
- Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, ON M5S 1A8, Canada
| | - Clare Fiala
- Laboratory Medicine Program, Department of Pathology, University Health Network, 200 Elizabeth Street, Toronto, ON M5G 2C4, Canada
| | - Alessia Portante
- Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, ON M5S 1A8, Canada
| | - Madhu Rabindranath
- Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, ON M5S 1A8, Canada
| | - Noor Alsafwani
- Laboratory Medicine Program, Department of Pathology, University Health Network, 200 Elizabeth Street, Toronto, ON M5G 2C4, Canada
- Department of Pathology, College of Medicine, Imam Abdulrahman Bin Faisal University, P.O. Box.2208, Dammam, 31441, Saudi Arabia
| | - Andrew Gao
- Laboratory Medicine Program, Department of Pathology, University Health Network, 200 Elizabeth Street, Toronto, ON M5G 2C4, Canada
- Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, ON M5S 1A8, Canada
| | - Ugljesa Djuric
- Princess Margaret Cancer Centre, 101 College Street, Toronto, ON M5G 1L7, Canada
| | - Phedias Diamandis
- Laboratory Medicine Program, Department of Pathology, University Health Network, 200 Elizabeth Street, Toronto, ON M5G 2C4, Canada
- Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, ON M5S 1A8, Canada
- Princess Margaret Cancer Centre, 101 College Street, Toronto, ON M5G 1L7, Canada
- Department of Medical Biophysics, University of Toronto, 101 College St, Toronto, ON M5G 1L7, Canada
| |
Collapse
|
14
|
Dehkharghanian T, Rahnamayan S, Riasatian A, Bidgoli AA, Kalra S, Zaveri M, Babaie M, Seyed Sajadi MS, Gonzalelz R, Diamandis P, Pantanowitz L, Huang T, Tizhoosh HR. Selection, Visualization, and Interpretation of Deep Features in Lung Adenocarcinoma and Squamous Cell Carcinoma. THE AMERICAN JOURNAL OF PATHOLOGY 2021; 191:2172-2183. [PMID: 34508689 DOI: 10.1016/j.ajpath.2021.08.013] [Citation(s) in RCA: 5] [Impact Index Per Article: 1.7] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Received: 02/25/2021] [Revised: 08/09/2021] [Accepted: 08/20/2021] [Indexed: 12/18/2022]
Abstract
Although deep learning networks applied to digital images have shown impressive results for many pathology-related tasks, their black-box approach and limitation in terms of interpretability are significant obstacles for their widespread clinical utility. This study investigates the visualization of deep features (DFs) to characterize two lung cancer subtypes, adenocarcinoma and squamous cell carcinoma. This study demonstrates that a subset of DFs exist that can accurately distinguish these two cancer subtypes, prominent DFs. Visualization of such individual DFs allows us to understand better histopathologic patterns at both the whole-slide and patch levels, allowing discrimination of these cancer types. These DFs were visualized at the whole slide image level through DF-specific heatmaps and at tissue patch level through generating activation maps. In addition, we show that these prominent DFs contain information that can distinguish carcinomas of organs other than the lung. This framework may serve as a platform for evaluating the interpretability of any deep network for diagnostic decision making.
Collapse
Affiliation(s)
- Taher Dehkharghanian
- Nature Inspired Computer Intelligence (NICI) Lab, Ontario Tech University, Oshawa, Ontario, Canada; Department of Pathology and Molecular Medicine, McMaster University, Hamilton, Ontario, Canada
| | - Shahryar Rahnamayan
- Nature Inspired Computer Intelligence (NICI) Lab, Ontario Tech University, Oshawa, Ontario, Canada
| | - Abtin Riasatian
- KIMIA (Laboratory for Knowledge Inference in Medical Image Analysis) Lab, University of Waterloo, Waterloo, Ontario, Canada
| | - Azam A Bidgoli
- Nature Inspired Computer Intelligence (NICI) Lab, Ontario Tech University, Oshawa, Ontario, Canada
| | - Shivam Kalra
- KIMIA (Laboratory for Knowledge Inference in Medical Image Analysis) Lab, University of Waterloo, Waterloo, Ontario, Canada
| | - Manit Zaveri
- KIMIA (Laboratory for Knowledge Inference in Medical Image Analysis) Lab, University of Waterloo, Waterloo, Ontario, Canada
| | - Morteza Babaie
- KIMIA (Laboratory for Knowledge Inference in Medical Image Analysis) Lab, University of Waterloo, Waterloo, Ontario, Canada
| | - Mahjabin S Seyed Sajadi
- KIMIA (Laboratory for Knowledge Inference in Medical Image Analysis) Lab, University of Waterloo, Waterloo, Ontario, Canada
| | - Ricardo Gonzalelz
- KIMIA (Laboratory for Knowledge Inference in Medical Image Analysis) Lab, University of Waterloo, Waterloo, Ontario, Canada
| | - Phedias Diamandis
- Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, Ontario, Canada
| | - Liron Pantanowitz
- Department of Pathology, University of Michigan, Ann Arbor, Michigan
| | - Tao Huang
- Department of Pathology, University of Michigan, Ann Arbor, Michigan
| | - Hamid R Tizhoosh
- KIMIA (Laboratory for Knowledge Inference in Medical Image Analysis) Lab, University of Waterloo, Waterloo, Ontario, Canada.
| |
Collapse
|
15
|
Yu P, Kibbe W. Cancer Data Science and Computational Medicine. JCO Clin Cancer Inform 2021; 5:487-489. [PMID: 33950710 DOI: 10.1200/cci.21.00006] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/25/2022] Open
Affiliation(s)
- Peter Yu
- Hartford Healthcare Cancer Institute, Hartford, CT
| | | |
Collapse
|
16
|
Tizhoosh HR, Diamandis P, Campbell CJV, Safarpoor A, Kalra S, Maleki D, Riasatian A, Babaie M. Searching Images for Consensus: Can AI Remove Observer Variability in Pathology? THE AMERICAN JOURNAL OF PATHOLOGY 2021; 191:1702-1708. [PMID: 33636179 DOI: 10.1016/j.ajpath.2021.01.015] [Citation(s) in RCA: 5] [Impact Index Per Article: 1.7] [Reference Citation Analysis] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Received: 10/08/2020] [Revised: 01/05/2021] [Accepted: 01/25/2021] [Indexed: 02/07/2023]
Abstract
One of the major obstacles in reaching diagnostic consensus is observer variability. With the recent success of artificial intelligence, particularly the deep networks, the question emerges as to whether the fundamental challenge of diagnostic imaging can now be resolved. This article briefly reviews the problem and how eventually both supervised and unsupervised AI technologies could help to overcome it.
Collapse
Affiliation(s)
| | - Phedias Diamandis
- Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, Canada
| | - Clinton J V Campbell
- Department of Pathology and Molecular Medicine, Faculty of Health Sciences, McMaster University, Hamilton, Ontario, Canada
| | - Amir Safarpoor
- Kimia Laboratory, University of Waterloo, Waterloo, Canada
| | - Shivam Kalra
- Kimia Laboratory, University of Waterloo, Waterloo, Canada
| | - Danial Maleki
- Kimia Laboratory, University of Waterloo, Waterloo, Canada
| | | | - Morteza Babaie
- Kimia Laboratory, University of Waterloo, Waterloo, Canada
| |
Collapse
|