timeline of medical artificial intelligence development

see also:

• timeline of artificial intelligence development and related issues
• artificial intelligence
• information technology main index
• AI Large Language Models (LLMs)
• https://health.google/intl/ALL_uk/health-research/imaging-and-diagnostics/

• mRNA signatures of immune response ¹⁾ leading to Inflammatix's MyRNA point of care device for precision medicine, and InSep acute infection and sepsis POC test which from a blood sample rapidly gives probability of bacterial vs viral infection as well as a severity value achieved using machine learning of transcriptome 29 mRNA signatures for thousands of classified patient blood samples with the challenge of mitigating ML over-fitting - version 3 now has 99% sens, 94% spec for bact. infection, 94% sens, 93% spec for viral infection, 94% sens, 93% spec for severity on ED and ICU patients ²⁾

• Measuring SaO2 using finger on smartphone camera and LED light
• measuring BP from forehead images

• Sybil: an accurate future prediction of lung cancer from one CT scan ³⁾
• coronary angiography ⁴⁾
• Dx endometriosis using US and MRI images ⁵⁾
• CXR - AI estimated “biologic age” - trained on over 100,000 CXRs ⁶⁾
• New AI Tool Beats Standard Approaches for Detecting Heart Attacks including HEART score ⁷⁾
• Interpretable Machine Learning on Metabolomics Data Reveals Biomarkers for Parkinson’s Disease ⁸⁾
• Ann Emerg Med. 2023 A Natural Language Processing and Machine Learning Approach to Identification of Incidental Radiology Findings in Trauma Patients Discharged from the Emergency Department
• Gothenburg Cardiac Arrest Machine Learning Studies - 30 day survival SCARS-1 OOHCA predictions are 95% sens and 89% spec with AUC-ROC value at 0.97
• https://www.news-medical.net/news/20230305/Causal-AI-quantifies-benefits-of-actions-needed-to-overcome-inherited-risk-of-coronary-artery-disease.aspx
• 2023: Nature "A deep learning algorithm to predict risk of pancreatic cancer from disease trajectories"
• UAE-based company, G42, is planning on using Cerebras' new 7nm 300mm wafer chip 1.5MW, 4exaflop 850,000 core WSE-2 supercomputer (each can process some 4 billion parameters and is some 200x faster than a nVidia A100 GPU system) to develop medical AGI using their vast medical databases
• Lancet 2023: Identification and risk stratification of coronary disease by artificial intelligence-enabled ECG
  • ECG-AI for identifying CAC ≥300, obstructive CAD, and regional akinesis achieved area under the receiver operating characteristic (AUROC) values of 0.88, 0.85, and 0.94, respectively. An ensembled ECG-AI identified 3, 5, and 10-year risk for acute coronary events and mortality independently and additively to PCE.
• Queensland company Remote Medical Innovations (RMI) Oceania develops a remote controlled robot for performing ultrasound such as echocardiography ⁹⁾

• Google announces AMIE new medical AI LLM which asks user for more information and outperforms doctors in diagnosis and empathy
• https://litfl.com/applications-of-artificial-intelligence-in-medicine/
• Neuralink implants the 1st wireless brain-computer interface using a special robot to surgically place the implants’ “ultra-fine” threads that help transmit signals in participants’ brains.
• Health-LLM framework proposed ¹⁰⁾
• nVidia partners with Hippocratic AI to produce digital AI video agent for nurse help line calls at $US9/hr cost which could replace nurses for such video calls and are said to out-perform human nurses on bedside manner, education and only narrowly misses on patient satisfaction measure
• DrugGPT to help UK prescribers
• Google DeepMind's Med-Gemini multi-modal LLM can read CXRs, perform needle-in-a-haystack searches within EHRs and outperforms GPT4 ¹¹⁾
• AI can detect pathology on CXRs (in terms of “remarkable” or “unremarkable”) as well as a radiologist it seems, however, the mistakes made by AI were, on average, more clinically severe for the patient than mistakes made by radiologists, presumably as AI was blinded to the clinical scenario whereas radiologists were not ¹²⁾
• in 2024, researchers, by leveraging cutting-edge proteogenomics and comparative genomics techniques to comprehensively annotate more than 964 million non-redundant smORFs across 75 habitats, a scale approximately ~20-fold greater than any previous smORF work by analyzing data from more than 63,000 metagenomes and almost 88,000 isolate genomes to construct a novel global microbial small open reading frames (smORFs) catalog (GMSC) ¹³⁾