MaltSci 麦伴科研

Appearance

Curated Papers > Highly Cited Authoritatives in "AI Drug Discovery" Literature

Artificial intelligence in digital pathology - new tools for diagnosis and precision oncology. ​

Literature Information ​

DOI10.1038/s41571-019-0252-y
PMID31399699
JournalNature reviews. Clinical oncology
Impact Factor82.2
JCR QuartileQ1
Publication Year2019
Times Cited605
KeywordsArtificial Intelligence, Digital Pathology, Precision Oncology, Biomarkers, Deep Learning
Literature TypeJournal Article, Research Support, N.I.H., Extramural, Research Support, Non-U.S. Gov't, Research Support, U.S. Gov't, Non-P.H.S., Review
ISSN1759-4774
Pages703-715
Issue16(11)
AuthorsKaustav Bera, Kurt A Schalper, David L Rimm, Vamsidhar Velcheti, Anant Madabhushi

TL;DR ​

This study highlights the growing need for predictive assays in precision oncology, emphasizing the complexity of cancer's signaling networks that hinder the development of effective biomarkers. It evaluates AI and machine learning approaches in digital pathology for mining morphometric features from tissue images, addressing challenges and future opportunities in enhancing patient management and biomarker development.

Search for more papers on MaltSci.com ​

Artificial Intelligence · Digital Pathology · Precision Oncology · Biomarkers · Deep Learning

Abstract ​

In the past decade, advances in precision oncology have resulted in an increased demand for predictive assays that enable the selection and stratification of patients for treatment. The enormous divergence of signalling and transcriptional networks mediating the crosstalk between cancer, stromal and immune cells complicates the development of functionally relevant biomarkers based on a single gene or protein. However, the result of these complex processes can be uniquely captured in the morphometric features of stained tissue specimens. The possibility of digitizing whole-slide images of tissue has led to the advent of artificial intelligence (AI) and machine learning tools in digital pathology, which enable mining of subvisual morphometric phenotypes and might, ultimately, improve patient management. In this Perspective, we critically evaluate various AI-based computational approaches for digital pathology, focusing on deep neural networks and 'hand-crafted' feature-based methodologies. We aim to provide a broad framework for incorporating AI and machine learning tools into clinical oncology, with an emphasis on biomarker development. We discuss some of the challenges relating to the use of AI, including the need for well-curated validation datasets, regulatory approval and fair reimbursement strategies. Finally, we present potential future opportunities for precision oncology.

MaltSci.com AI Research Service ​

Intelligent ReadingAnswer any question about the paper and explain complex charts and formulas
Locate StatementsFind traces of a specific claim within the paper
Add to KBasePerform data extraction, report drafting, and advanced knowledge mining

Primary Questions Addressed ​

  1. What are the specific challenges faced in developing functionally relevant biomarkers using AI in digital pathology?
  2. How do different AI methodologies, such as deep neural networks versus hand-crafted features, compare in their effectiveness for biomarker development?
  3. What role do well-curated validation datasets play in the implementation of AI tools in clinical oncology?
  4. How can regulatory approval processes impact the adoption of AI technologies in digital pathology?
  5. What future advancements in AI could further enhance precision oncology and patient management strategies?

Key Findings ​

Background and Purpose ​

In recent years, the integration of artificial intelligence (AI) into digital pathology has emerged as a promising avenue for enhancing diagnostic accuracy and patient management in precision oncology. The study aims to evaluate various AI-based computational methods, particularly focusing on deep learning and hand-crafted feature approaches, to facilitate the development of functional biomarkers that can assist in patient stratification and treatment selection.

Main Methods/Materials/Experimental Design ​

The authors present a structured workflow for the implementation of AI in digital pathology, detailing the steps involved in digitizing tissue specimens and applying AI algorithms for image analysis. The methods include:

  1. Tissue Preparation: Biopsy or surgical resection of specimens, followed by fixation and sectioning.
  2. Whole-Slide Imaging: Use of whole-slide scanners to digitize the tissue slides.
  3. Image Analysis:
    • Deep Learning (DL): Utilizing convolutional neural networks (CNNs) for automatic detection and classification of tissue features.
    • Hand-Crafted Features: Developing specific morphological features based on domain knowledge to enhance predictive accuracy.
  4. Performance Evaluation: Comparing AI predictions against expert pathologist assessments to determine diagnostic accuracy.
Mermaid diagram

Key Results and Findings ​

  • AI methods have shown diagnostic performance comparable to that of expert pathologists, with significant accuracy in detecting and classifying various cancer types.
  • The integration of AI in pathology workflows has the potential to reduce human error rates and improve the efficiency of diagnostic processes.
  • Studies indicate that AI can enhance the identification of prognostic biomarkers, leading to better patient stratification and tailored treatment plans.

Main Conclusions/Significance/Innovation ​

The integration of AI in digital pathology represents a transformative shift in oncology diagnostics. By automating image analysis and providing robust predictive capabilities, AI tools can assist pathologists in overcoming subjective biases and improve diagnostic consistency. This innovation is expected to facilitate the development of multi-dimensional biomarkers that reflect the complex biology of tumors, thereby advancing precision medicine.

Research Limitations and Future Directions ​

  • Data Quality: The performance of AI algorithms heavily relies on the quality and quantity of annotated training data. Efforts must be made to curate high-quality datasets.
  • Regulatory Challenges: The approval process for AI-based diagnostic tools remains complex, necessitating clear regulatory pathways.
  • Interpretability: Many deep learning models operate as "black boxes," which can hinder clinical acceptance. Future research should focus on enhancing model interpretability.
  • Integration with Clinical Practice: Establishing reimbursement models and integrating AI tools into routine clinical workflows will be crucial for widespread adoption.

Future directions include the exploration of multimodal approaches that combine histological, genomic, and proteomic data to create comprehensive diagnostic tools that can better inform treatment decisions. Additionally, advancements in imaging technologies may pave the way for new AI applications in digital pathology.

References ​

  1. Gene expression and benefit of chemotherapy in women with node-negative, estrogen receptor-positive breast cancer. - Soonmyung Paik;Gong Tang;Steven Shak;Chungyeul Kim;Joffre Baker;Wanseop Kim;Maureen Cronin;Frederick L Baehner;Drew Watson;John Bryant;Joseph P Costantino;Charles E Geyer;D Lawrence Wickerham;Norman Wolmark - Journal of clinical oncology : official journal of the American Society of Clinical Oncology (2006)
  2. Systematic analysis of breast cancer morphology uncovers stromal features associated with survival. - Andrew H Beck;Ankur R Sangoi;Samuel Leung;Robert J Marinelli;Torsten O Nielsen;Marc J van de Vijver;Robert B West;Matt van de Rijn;Daphne Koller - Science translational medicine (2011)
  3. Validating whole slide imaging for diagnostic purposes in pathology: guideline from the College of American Pathologists Pathology and Laboratory Quality Center. - Liron Pantanowitz;John H Sinard;Walter H Henricks;Lisa A Fatheree;Alexis B Carter;Lydia Contis;Bruce A Beckwith;Andrew J Evans;Avtar Lal;Anil V Parwani; - Archives of pathology & laboratory medicine (2013)
  4. A Deep Convolutional Neural Network for segmenting and classifying epithelial and stromal regions in histopathological images. - Jun Xu;Xiaofei Luo;Guanhao Wang;Hannah Gilmore;Anant Madabhushi - Neurocomputing (2016)
  5. Classification and mutation prediction from non-small cell lung cancer histopathology images using deep learning. - Nicolas Coudray;Paolo Santiago Ocampo;Theodore Sakellaropoulos;Navneet Narula;Matija Snuderl;David Fenyö;Andre L Moreira;Narges Razavian;Aristotelis Tsirigos - Nature medicine (2018)
  6. A multigene assay to predict recurrence of tamoxifen-treated, node-negative breast cancer. - Soonmyung Paik;Steven Shak;Gong Tang;Chungyeul Kim;Joffre Baker;Maureen Cronin;Frederick L Baehner;Michael G Walker;Drew Watson;Taesung Park;William Hiller;Edwin R Fisher;D Lawrence Wickerham;John Bryant;Norman Wolmark - The New England journal of medicine (2004)
  7. Automated image analysis system for detecting boundaries of live prostate cancer cells. - I Simon;C R Pound;A W Partin;J Q Clemens;W A Christens-Barry - Cytometry (1998)
  8. Artificial Intelligence and Digital Pathology: Challenges and Opportunities. - Hamid Reza Tizhoosh;Liron Pantanowitz - Journal of pathology informatics (2018)
  9. PD-L1 biomarker testing for non-small cell lung cancer: truth or fiction? - Claud Grigg;Naiyer A Rizvi - Journal for immunotherapy of cancer (2016)
  10. Interobserver variability between expert urologic pathologists for extraprostatic extension and surgical margin status in radical prostatectomy specimens. - Andrew J Evans;Pauline C Henry;Theodorus H Van der Kwast;Douglas C Tkachuk;Kemp Watson;Gina A Lockwood;Neil E Fleshner;Carol Cheung;Eric C Belanger;Mahul B Amin;Liliane Boccon-Gibod;David G Bostwick;Lars Egevad;Jonathan I Epstein;David J Grignon;Edward C Jones;Rodolfo Montironi;Madeleine Moussa;Joan M Sweet;Kiril Trpkov;Thomas M Wheeler;John R Srigley - The American journal of surgical pathology (2008)

Literatures Citing This Work ​

  1. Spatially multiplexed RNA in situ hybridization to reveal tumor heterogeneity. - Lena Voith von Voithenberg;Anna Fomitcheva Khartchenko;Deborah Huber;Peter Schraml;Govind V Kaigala - Nucleic acids research (2020)
  2. Deep learning pathological microscopic features in endemic nasopharyngeal cancer: Prognostic value and protentional role for individual induction chemotherapy. - Kuiyuan Liu;Weixiong Xia;Mengyun Qiang;Xi Chen;Jia Liu;Xiang Guo;Xing Lv - Cancer medicine (2020)
  3. Opportunities for Artificial Intelligence in Advancing Precision Medicine. - Fabian V Filipp - Current genetic medicine reports (2019)
  4. Introduction to digital pathology and computer-aided pathology. - Soojeong Nam;Yosep Chong;Chan Kwon Jung;Tae-Yeong Kwak;Ji Youl Lee;Jihwan Park;Mi Jung Rho;Heounjeong Go - Journal of pathology and translational medicine (2020)
  5. Glioma Grading via Analysis of Digital Pathology Images Using Machine Learning. - Saima Rathore;Tamim Niazi;Muhammad Aksam Iftikhar;Ahmad Chaddad - Cancers (2020)
  6. Computationally Derived Image Signature of Stromal Morphology Is Prognostic of Prostate Cancer Recurrence Following Prostatectomy in African American Patients. - Hersh K Bhargava;Patrick Leo;Robin Elliott;Andrew Janowczyk;Jon Whitney;Sanjay Gupta;Pingfu Fu;Kosj Yamoah;Francesca Khani;Brian D Robinson;Timothy R Rebbeck;Michael Feldman;Priti Lal;Anant Madabhushi - Clinical cancer research : an official journal of the American Association for Cancer Research (2020)
  7. Image-guided tumor surgery: The emerging role of nanotechnology. - Nicholas E Wojtynek;Aaron M Mohs - Wiley interdisciplinary reviews. Nanomedicine and nanobiotechnology (2020)
  8. Editorial: Digital Interventions in Mental Health: Current Status and Future Directions. - Elias Aboujaoude;Lina Gega;Michelle B Parish;Donald M Hilty - Frontiers in psychiatry (2020)
  9. Artificial intelligence driven next-generation renal histomorphometry. - Briana A Santo;Avi Z Rosenberg;Pinaki Sarder - Current opinion in nephrology and hypertension (2020)
  10. Emerging role of deep learning-based artificial intelligence in tumor pathology. - Yahui Jiang;Meng Yang;Shuhao Wang;Xiangchun Li;Yan Sun - Cancer communications (London, England) (2020)

... (595 more literatures)

© 2025 MaltSci - We reshape scientific research with AI technology