|Name||Position||Country||Term||Time in Office|
|LJ Kricka||Chair||US||1st||2021 01 - 2023 12|
The IFCC Emerging Technology Division (ETD) is dedicated to providing current awareness for emerging technologies likely to have important clinical diagnostic applications in the near future. Artificial intelligence (AI) is an important emerging technology, and in the future, it is likely that this technology will become embedded in many aspects of medicine. In particular, it is expected to play a key role in laboratory medicine. Accordingly, a familiarity with this technology and its scope, applications, accessibility, and limitations will become important in the practice of laboratory medicine in the future. The focus of this WG is the area defined by the intersection of artificial intelligence, genomics, and clinical diagnostics.
Q: “What are the applications of AI in genomic diagnostics?”
A: “Artificial intelligence (AI) has many potential applications in genomic diagnostics, which involve the use of genomic information to identify genetic conditions and guide treatment decisions. Some examples of how AI might be used in this field include:
Analyzing large amounts of genomic data: AI algorithms can be used to analyze large amounts of genomic data quickly and accurately, helping to identify genetic variations and patterns that may be relevant to a particular condition or disease.
Predicting disease risk: AI can be used to predict an individual's risk of developing a particular disease based on their genomic information, allowing for earlier diagnosis and treatment.
Personalized medicine: AI can be used to tailor treatment plans to an individual's specific genomic profile, helping to optimize treatment effectiveness and minimize side effects.
Drug discovery: AI can be used to identify potential new drugs and evaluate their effectiveness based on an individual's genomic information, helping to speed up the drug development process.
Clinical decision support: AI can be used to provide clinicians with recommendations for diagnosis and treatment based on an individual's genomic information, helping to improve the accuracy and efficiency of healthcare delivery.”
The above summary of the specific application area of interest of this WG was generated by ChatGPT a new AI-based chatbot from OpenAI (openai.com). I will let the chatbot explain what exactly it is by simply asking it!
Q: “What is ChatGPT?”
A: “ChatGPT is a chatbot based on the GPT-3 language model, developed by OpenAI. GPT-3 (short for "Generative Pre-training Transformer 3") is a state-of-the-art natural language processing (NLP) model that has been trained on a vast amount of data and can generate human-like text when given a prompt. ChatGPT is designed to be able to hold a conversation with a human user in real-time, responding to user input with appropriate and coherent responses.
To use ChatGPT, a user can type a message into a chat window and ChatGPT will generate a response based on the message it was given. The user can then continue the conversation by typing a follow-up message, and ChatGPT will generate a response to that as well. This process can continue for multiple turns, allowing for a back-and-forth conversation between the user and the chatbot.
ChatGPT is an example of how AI and NLP technology can be used to build chatbots that are able to hold natural and engaging conversations with humans. These chatbots can be used in a variety of applications, such as customer service, online tutoring, and entertainment.”
What it did not tell you is that it is not currently connected to the internet and has only limited information after 2021 (it was trained using 570 GB of data sourced from various types of content [e.g., books, Wikipedia, websites] comprising 300 billion words), so we are only seeing the very beginning of the utility of this AI-based technology (see https://analyticsindiamag.com/behind-chatgpts-wisdom-300-bn-words-570-gb-data/, https://www.nytimes.com/2022/12/05/technology/chatgpt-ai-twitter.html, https://www.medrxiv.org/content/10.1101/2022.12.19.22283643v1.full.pdf, https://www.nature.com/articles/d41586-022-04437-2, https://www.medrxiv.org/content/10.1101/2022.12.16.22283512v2).
NEW! Recommendations for data, source code, AI methods, and experiments in AI publications (adapted from Gundersen et al 2018) can be found HERE.
NEW! Several recent reports and publications addressing the convergence of digital health, artificial intelligence with Universal Health Coverage (The Lancet and Financial Times Commission), ethics and governance of artificial intelligence (WHO), good machine learning practice (FDA), and the reproducibility and clinical translation of machine learning applications can be found BELOW:
NEW! 2022 AI Index Report (now in its Fifth Edition) (Daniel Zhang, Nestor Maslej, Erik Brynjolfsson, John Etchemendy, Terah Lyons, James Manyika, Helen Ngo, Juan Carlos Niebles, Michael Sellitto, Ellie Sakhaee, Yoav Shoham, Jack Clark, and Raymond Perrault, “The AI Index 2022 Annual Report,” AI Index Steering Committee, Stanford Institute for Human-Centered AI, Stanford University, March 2022. https://hai.stanford.edu/research/ai-index-2022).
This annual report “tracks, collates, distills, and visualizes data relating to artificial intelligence, enabling decision-makers to take meaningful action to advance AI responsibly and ethically with humans in mind.” The report is divided into five chapters covering: Research and Development, Technical Performance, Technical AI Ethics, The Economy and Education, AI Policy and Governance.
Useful commentaries on the report can be found in a series of articles on: a. the state of AI in 9 charts (Lynch S. Stanford University HAI. Mar 16, 2022. https://hai.stanford.edu/news/state-ai-9-charts); b. the mounting ethical concerns (Zhang D, Clark J, Perrault R. The 2022 AI Index: Industrialization of AI and Mounting Ethical Concerns. Stanford University HAI. Mar 16, 2022. https://hai.stanford.edu/news/2022-ai-index-industrialization-ai-and-mounting-ethical-concerns); and c. AIs ethical growing pains (Miller K. The 2022 AI Index: AI’s Ethical Growing Pains. Stanford University HAI. Mar 16, 2022. https://hai.stanford.edu/news/2022-ai-index-ais-ethical-growing-pains).
NEW! 2022 Current Awareness Update This Current Awareness Update covers the first part of 2022. The continuing importance of AI in genomics is underscored by a recent market analysis that predicts that the artificial intelligence in genomics market will grow at a compound annual growth rate (CAGR) of over 48% in the next five years (2022-2027) (see https://www.researchandmarkets.com/reports/5544413/artificial-intelligence-in-genomics-market).
Recent AI in genomics literature covers machine learning in cancer (1), Covid-19 (2), variants of uncertain significance (3), genetic relatedness (4), Personalized Medicine (5), disease progression (6), viral and glioblastoma diagnostics (7,8), variant classification and prioritization (9), precision medicine (10), and biomarker assessment (11). Two recent books have provided insights into deep learning and machine learning in clinical and genomic diagnostics (12) and machine learning methods in the classification of genomics data (13).
Particularly relevant to an emerging technology such as AI in genomics is a recent review that examines and illustrates the pitfalls of applying supervised machine learning in genomics (14). Also noteworthy is a recent update (January 2022) of the National Human Genome Research Institute web page on “Artificial intelligence, machine learning and genomics” (NHGRI) (15). This informative web page addresses the topic via a series of questions (e.g., “What are some ways in which AI/ML are being used in genomics?).
NEW! 2022 Current Awareness Update. Regulatory landscape of AI/ML- Enabled Medical Devices
Recent (2021-2) documents published by regulatory and government bodies relating to AI/ML enabled devices and Software as a Medical Device (SaMD) together with reviews and commentaries can be found HERE.
Prof Larry KRICKA
Department of Pathology and Laboratory Medicine
University of Pennsylvania Medical Center
3400 Spruce Street
Philadelphia, PA 19104 - USA