Artificial Intelligence in Pathology

Sections on this page:

• Current News
• Definitions of Artificial Intelligence and Machine Learning
• Concept of Augmented Intelligence
• How AI/ML may be used in Pathology
• Ethical use of AI in Healthcare
• Regulation of Artificial Intelligence and Machine Learning
• CAP Activities

Multimodal data integration to predict patient outcomes
September 6, 2022

Use of multimodal data, including histopathological, genomic, and transcriptomic heterogeneity in analyzing tumor and tissue microenvironment may show insights into patient prognostication further than what is currently capable today with conventional outcome predictions. In this study, investigators used a deep learning architecture to train a machine learning model to predict patient outcomes from whole slide images derived from histology tissue and genomic sequencing. The study investigators used paired patient (n=5,720) data including 6,592 whole slide images and molecular data from The Cancer Genome Atlas to train and validate a weakly supervised machine learning model using 5-fold cross-validation and compare it to Cox models with clinical variables and unimodal deep-learning models. Their method outperformed these models on 12 of the 14 cancer types in a one-versus-all comparison. A goal of this research was to develop an interpretable, weakly-supervised, multimodal deep learning algorithm that integrates whole slide images (WSIs) and molecular profile features for cancer prognosis. Human evaluation of the attention maps showed that regions with higher nuclear:cytoplasmic ratio and high tumor infiltrating lymphocytes weighted heavily as prognostic features. The model was also able to independently identify IDH1 mutations as a prognostic factor in low grade gliomas. The researched published their code named, PORPOISE, a freely accessible interactive application that uses their model to immediately produce WSI and molecular feature explanations for each of the 14 cancer types. They also provide a demo. This shows significant proof-of-concept performance where machine learning models may infer good prognostication for patients compared to the current paradigm.

1. Chen, Richard J. et al. (2022). Pan-cancer integrative histology-genomic analysis via multimodal deep learning. Cancer Cell, Volume 40, Issue 8, 865 – 878.e6 https://www.cell.com/cancer-cell/fulltext/S1535-6108(22)00317-8

Prostate Cancer Risk Stratification via light sheet microscopy
December 1, 2021

Prostate cancer diagnostics are heavily reliant on pathologist interpretation of thin 2D sections of prostate biopsies. These sections are used to infer the 3D structure of the cancer and classify using the ISUP grading system, which is correlated with patient outcomes and used to make high impact clinical decisions. ISUP grading has known challenges with inter-pathologist variation. This variability can lead to misclassification of patients and both over- and undertreatment of their disease. This article presents a new approach to imaging prostate biopsies in 3D and using AI to predict recurrence of the patient's cancer. The results show that 3D pathology with AI is accurate in stratifying high risk versus low risk patients, and outperforms AI approaches using traditional 2D sections. The study is notable for its use of an ex-vivo microscopy technique (light-sheet microscopy), 3D imaging of entire prostate core needle biopsies, and the application of AI to 3D digital pathology datasets from clinical biopsies. If validated in larger patient cohorts, the technology presents a promising new prostate cancer diagnostic to be used side-by-side with pathologist interpretation of traditional 2D sections.

1. Xie W, Reder NP, Koyuncu C, Leo P, Hawley S, Huang H, Mao C, Postupna N, Kang S, Serafin R, Gao G, Han Q, Bishop KW, Barner LA, Fu P, Wright JL, Keene CD, Vaughan JC, Janowczyk A, Glaser AK, Madabhushi A, True LD, Liu JTC. Prostate Cancer Risk Stratification via Nondestructive 3D Pathology with Deep Learning-Assisted Gland Analysis. Cancer Res. 2022 Jan 15;82(2):334-345. doi: 10.1158/0008-5472.CAN-21-2843. Epub 2021 Dec 1. PMID: 34853071; PMCID: PMC8803395.

"Rise of the machines" AI is not just for AP anymore
September 24, 2021

Laboratory medicine has continued to generate paramount data related to medical decision making. Much of the machine learning/artificial intelligence literature has circulated digital imaging in pathology, however a recently published review has outlined the various use cases where ML/AI can be leveraged in clinical pathology. While image analysis and rare event detection may also be prevalent in lab medicine, other applications also include applications include instrument automation, error detection, forecasting, result interpretation, test utilization, and genomics. The authors also cover the challenges that exist related to machine learning in healthcare and laboratory medicine. In addition, discussion of a proof of concept related to automated machine learning.

1. The Journal of Applied Laboratory Medicine, jfab075, https://doi.org/10.1093/jalm/jfab075
2. Rashidi, HH, Tran, N, Albahra, S, Dang, LT. Machine learning in health care and laboratory medicine: General overview of supervised learning and Auto-ML. Int J Lab Hematol. 2021; 43: 15– 22. https://doi.org/10.1111/ijlh.13537

First FDA cleared AI product in Digital Pathology
September 21, 2021

The first marketing authorization for an AI product in digital pathology has been given by FDA to Paige for its prostate cancer detection device, Paige Prostate. This software is aimed to assist pathologists in the detection of areas that are suspicious for cancer as an adjunct to the review of digitally scanned whole slide images (WSIs) derived from prostate biopsies. The FDA has also created a new product classification, “Software algorithm device to assist users in digital pathology,” and has described this generic type of device as; A software algorithm device to assist users in digital pathology is an in vitro diagnostic device intended to evaluate acquired scanned pathology whole slide images. The device uses software algorithms to provide information to the user about presence, location, and characteristics of areas of the image with clinical implications. Information from this device is intended to assist the user in determining a pathology diagnosis.

The FDA has granted this first de novo for an AI product by evaluating data from a clinical study where 16 pathologists examined 527 WSIs of prostate needle core biopsies (171 cancer and 356 benign) that were digitized using a scanner. For each WSI, the pathologists completed two assessments, one without Paige Prostate’s assistance (unassisted read) and one with Paige Prostate’s assistance (assisted read). The study found that Paige Prostate improved detection of cancer on individual WSIs by 7.3% on average when compared to pathologists’ unassisted reads for WSIs of individual biopsies, with no impact on the read of benign WSIs. Paige Prostate also showed the largest pathologist sensitivity improvement on challenging small tumors (less than 0.4 mm), where their performance improved by 12.5% on average.

Read FDA News Announcement  

Ethics of Artificial Intelligence in Pathology and Laboratory Medicine
August 24, 2021

AI ethics, both within and outside of healthcare, is increasingly discussed in news headlines, journals, and conferences. The referenced article (mini-review) related to Ethics of artificial intelligence in pathology, makes two important contributions to this discourse. First, it addresses AI ethics specifically within the context of pathology and laboratory medicine. Second, it structures the discussion around the ethical traditions of human subject research and science more broadly. The principle of autonomy, also known as respect for persons, has traditionally meant that individuals could decide for themselves what should happen to their physical body. This is often addressed through informed consent, such as for surgical procedures and research studies. Many people believe that autonomy extends to an individual’s data as well, and not just their physical body. Beneficence and nonmaleficence (do no harm) mean that technologies must have a realistic expectation of benefit to the individual, along with a low risk of harm. Justice means that the risks and benefits (including financial benefits) should be distributed in ways that society considers to be fair. In addition to these human subject research principles of autonomy, beneficence, and justice, the paper also notes the scientific ethical tradition that prioritizes knowledge acquisition and open sharing. Finally, the paper describes some ways in which these principles can be enforced, not just through individual professional accountability, but also at an organizational level. Developers, implementers, and validation efforts using machine learning should ensure systems follow he discussed ethical principles.

1. Jackson BR, Ye Y, Crawford JM, Becich MJ, Roy S, Botkin JR, de Baca ME, Pantanowitz L. The Ethics of Artificial Intelligence in Pathology and Laboratory Medicine: Principles and Practice. Acad Pathol. 2021 Feb 16;8:2374289521990784. doi: 10.1177/2374289521990784. PMID: 33644301; PMCID: PMC7894680.

FDA Evaluations of Medical AI Devices Show Limitations
April 13, 2021

While there are no regulatory clearances for pathology related AI devices, this article stresses critical points in the limitations in several currently available FDA cleared medical AI devices. The authors also provide an annotated database of the 130 medial AI devices analyzed in their article, including risk level, demographic availability, and if multiple site data was evaluated (Database). The authors discuss two main limitations in development of AI models including 1) the use of only retrospective data when developing the model; and 2) the lack of generalizability from insufficient variation of data sources or inclusion of only a single/few sites in developing the model. The potential influence of human-computer-interaction in a prospective setting to deviate the model’s intended use should be evaluated in a prospective setting, such that a device cleared as a screening tool is not used as a primary diagnostic tool. Importantly, the use of varied demographic data is critical to ensure inclusion and evaluation of the model in diverse patient populations. The authors presented a case study evaluating three models trained on three publicly available chest x-ray datasets for pneumothorax detection. Each model was trained at a single site and tested across the three datasets. The model’s with the highest performance were the ones trained at the same site, and there was significant decreases in AUC for models trained at a different site than where the test data was evaluated. Recommendations include evaluating the performance of an AI device at multiple clinical sites, encouraging prospective evaluation during clinical trials, and post-market surveillance.

1. Wu, E., Wu, K., Daneshjou, R. et al. How medical AI devices are evaluated: limitations and recommendations from an analysis of FDA approvals. Nat Med (2021). https://doi.org/10.1038/s41591-021-01312-x

Artificial intelligence (AI) is the ability of computer software to mimic human judgement. Current AI systems carry out only very specific tasks for which they are designed, but they may integrate large amounts of input data to carry out these tasks quickly and accurately. The current excitement about AI is focused on machine learning (ML) systems and this domain is sometimes referred to as AI/ML. AI/ML systems may be trained using defined input data sets, which may include images, to associate patterns in data with clinical contexts such as diagnoses or outcomes. Once trained, AI/ML systems are used with new data to predict diagnosis or outcome in specific cases, or carry out other useful tasks. To date, systems are limited in the range of diagnoses, predictions, and tasks covered, but can be impressively accurate within their defined scope.

1.  Yu K-H, Beam AL, Kohane IS. Artificial intelligence in healthcare. Nature Biomedical Engineering. 2018;2:719-731.
2. Topol EJ. High-performance medicine: The convergence of human and artificial intelligence. Nature Medicine. 2019;25:44-56.

The American Medical Association has popularized the term Augmented Intelligence to represent the use of AI/ML as a tool to enhance rather than replace human healthcare providers. The Augmented Intelligence concept is based on studies that integrate AI/ML with human experts in a synergistic workflow that achieves higher performance than either separately. In the pathology context, Augmented Intelligence brings the computational advantages of AI/ML into the clinical and laboratory setting in the form of supportive tools that can enhance pathologists’ diagnostic capabilities by, for example, suggesting regions of interest or counting elements on a slide, or providing decision support to inform clinical judgement.

1. AMA. Artificial Intelligence in Medicine
2. Colling R, Pitman H, Oien K et al. Artificial intelligence in digital pathology: a roadmap to routine use in clinical practice. J Pathol. 2019;249:143-150.

Pathologists who are interested in AI/ML envision a variety of tools that may provide increased efficiency and diagnostic accuracy in the pathologist’s daily diagnostic workflow. As noted above, tools for the pathologist could scan slides to count elements such as lymph node metastases, mitoses, inflammatory cells, or pathologic organisms, presenting results at sign-out and flagging examples for review. AI/ML tools could also flag regions of interest on a slide or prioritize cases based on slide content. Studies to date have shown promise for automated detection of foci of cancer and invasion, tissue/cell quantification, virtual immunohistochemistry, spatial cell mapping of disease, novel staging paradigms for some types of tumors, and workload triaging. Future systems may be able to correlate patterns across multiple inputs from the medical record, including genomics, allowing a more comprehensive prognostic statement in the pathology report.

1. Pantanowitz L, Quiroga-Garza GM, Bien L et al. An artificial intelligence algorithm for prostate cancer diagnosis in whole slide images of core needle biopsies: a blinded clinical validation and deployment study. The Lancet Digital Health. 2020;2:e407-e416.
2. Colling R, Pitman H, Oien K et al. Artificial intelligence in digital pathology: a roadmap to routine use in clinical practice. J Pathol. 2019;249:143-150.
3. Campanella G, Hanna MG, Geneslaw L, et al. Clinical-grade computational pathology using weakly supervised deep learning on whole slide images. Nat Med. 2019;25(8):1301–1309.
4. Rashidi HH, Tran NK, Betts EV, Howell LP, Green R. Artificial intelligence and machine learning in pathology: The present landscape of supervised methods. Acad Pathol. 2019;6:2374289519873088.
5. Mezheyeuski A, Bergsland CH, Backman M, et al. Multispectral imaging for quantitative and compartment-specific immune infiltrates reveals distinct immune profiles that classify lung cancer patients. J Pathol. 2018;244(4):421–431.
6. Wilkes EH, Rumsby G, Woodward GM. Using Machine Learning to Aid the Interpretation of Urine Steroid Profiles. Clin Chem. 2018;64:1586-1595.
7. Arnaout R. Machine Learning in Clinical Pathology: Seeing the Forest for the Trees. Clin Chem. 2018;64(11):1553–1554.
8. Cabitza F, Banfi G. Machine learning in laboratory medicine: waiting for the flood?. Clin Chem Lab Med. 2018;56(4):516–524.

The need for large sets of patient data to train AI/ML algorithms raises issues of patient consent, privacy, data security, and data de-identification in the production of AI/ML systems. There is also an ethical duty to review algorithms prior to implementation and verify their performance at deployment to ensure that they are safe, efficacious, and reliable. Recent experience has shown that subtle biases may be incorporated into training data and influence the performance of the resulting systems; these must be mitigated and training data must reflect the diversity of the patient population that the AI/ML systems are intended to serve. An algorithm trained without using best practices for representing ethnic groups, socioeconomic classes, ages, and/ or sex may limit system generalizability to these patient populations in real world settings and exclude (or harm) these groups inadvertently. The “black box” nature of some popular algorithms (not revealing the data patterns associated with particular predictions) combined with the natural proprietary orientation of system vendors may lead to transparency problems and difficulty checking the algorithms by independent interpretation. Finally, the human resource toll of AI/ML must be considered: deskilling of the workforce through dependence on AI/ML must be mitigated and there will be a need to repurpose job roles to adapt to increasing automation.

1. Jackson BR, Ye Y, Crawford JM et al. The Ethics of Artificial Intelligence in Pathology and Laboratory Medicine: Principles and Practice. Academic Pathology. 2021;8:237428952199078.
2. Keskinbora KH. Medical ethics considerations on artificial intelligence. J Clin Neurosci. 2019;64:277-282.
3. O’Sullivan S, Nevejans N, Allen C et al. Legal, regulatory, and ethical frameworks for development of standards in artificial intelligence (AI) and autonomous robotic surgery. Int J Med Robot. 2019;15:e1968.

The training and use of AI/ML algorithms introduces a fundamentally new kind of data analysis into the healthcare workflow that requires an appropriate regulatory framework. By virtue of their influence on pathologists and other physicians in selection of diagnoses and treatments, the outputs of these algorithms can critically impact patient care. The data patterns identified by these systems are often not exact: there is not perfect separation of classes or predictions. Thus there are analogies with sensitivity, specificity, and predictive value of other complex tests performed by clinical laboratories. However, in machine learning the patterns in data are identified by software and often are not explicitly revealed. Biases or subtle errors may be incorporated inadvertently into machine learning systems and these must be identified and mitigated prior to deployment. Naturally occurring changes in healthcare context such as case mix changes, updated tests or sample preparation, or new therapies, may also change the input data profile and reduce the accuracy of a previously well-functioning machine learning system. 

An effective and equitable regulatory framework for machine learning in healthcare will 1) define requirements based on risk, i.e., tailored to the likelihood and magnitude of possible harm from each machine learning application, 2) require best practices for system development by vendors including bias assessment and mitigation, 3) define appropriate best practices for verification of system performance at deployment sites, i.e., local laboratories, 4) define best practices for monitoring the performance of machine learning systems over time and mitigating performance problems that develop, and 5) clearly assign responsibility for problems if and when they occur.

The development of this framework is in early stages. To date, the White House has released draft guidance for regulation of artificial intelligence applications that provides a set of high-level principles to which a regulatory framework in any domain should adhere. Specific to healthcare, the FDA has released proposals for processes leading to approval or clearance of machine learning software for use as a medical device. None of these proposals yet addresses best practices for local performance verification and monitoring of machine learning systems analogous to CLIA-mandated laboratory test performance requirements. The CAP regards this omission as a gap in current regulatory planning for machine learning in healthcare and is promoting the development of a more complete regulatory framework that will include guidance, approved methods, and best practices for local laboratories in deploying machine learning tools as they become available.

1.  Shulz WL, Durant TJS, Krumholz HM. Validation and regulation of clinical artificial intelligence. Clin Chem 2019;65:1336-1337.
2. Allen TC. Regulating artificial intelligence for a successful pathology future. Arch Pathol Lab Med 2019;143(10):1175.
3. Office of Management and Budget. Guidance for regulation of artificial intelligence applications. White House Memo. 2020;Jan 7:1-15.
4. FDA. Proposed regulatory framework for modifications to artificial intelligence/machine learning (AI/ML)-based software as a medical device (SaMD): Discussion paper and request for feedback. 2019;1-20.
5. FDA. Artificial Intelligence/Machine Learning (AI/ML)-Based Software as a Medical Device (SaMD) Action Plan. 2021; Jan 12:1-7.

The CAP is engaged in several activities targeting AI/ML. Internally, the Informatics Committee has formed a Machine Learning Working Group focused on education and technical issues particularly related to verification and performance monitoring. This group is sharing its technical work with the FDA. The Information Technology Leadership Committee has formed an AI Project Team to ensure coordination and alignment of AI/ML activities across the organization and to provide reports to the BOG. An AI in Anatomic Pathology Work Group, reporting to the Council on Scientific Affairs, is developing use cases for AI/ML in pathology that may evolve into PT programs.

Externally, the CAP participates in a several organizations including the Alliance for Digital Pathology, a collaborative group interested in the evolution of regulatory science as it applies to digital pathology and AI. The CAP also works with the American College of Radiology Data Science Institute, a resource in understanding how radiologists are developing and using AI systems. In addition, the CAP is the Primary Secretariat to the Integrating the Healthcare Enterprise’s International Pathology and Lab Medicine domain as well as DICOM Working Group 26: Pathology. These standards organizations are developing technical profiles for incorporation of AI/ML systems into healthcare that will be available to developers of AI/ML tools and systems.