AI In Healthcare Milestones Q3 2022
Margaretta Colangelo
Leading AI Analyst | Speaker | Writer | AI Newsletter 56,900+ subscribers
This is my new AI in Healthcare Highlights and Milestones report for Q3 2022. In this report I highlight clinically validated AI solutions across the healthcare spectrum including updates on several AI devices that have received regulatory clearance. I also provide updates on fundraising, partnerships, and overall growth in the AI ecosystem worldwide.
This report includes detailed analyses of AI applications that have been used successfully in the following areas: AI to sequence DNA to diagnose rare diseases, AI to predict atrial fibrillation, AI to improve diagnosis of heart disease, AI to discover a novel drug candidate to treat COVID-19, AI that accurately predicts if and when a person will experience cardiac arrest, AI that identifies hard to detect endoscopic kidney stones, and AI that accurately diagnoses early-stage Parkinson's disease for the first time.
"Another masterful work from Margaretta’s deep look into how AI and ML benefit our lives. Thank you, Margaretta for keeping us healthy."
Asian Robotics Review
This report also includes updates on several AI devices that have received regulatory clearance including:
Some of the companies and universities included in this report include Insilico Medicine, PathAI, iSono Health, Indica Labs, Gleamer, Aidoc, PreciseDx, Ibex Medical Analytics, Deepcell, BenevolentAI, Roche, Bristol Myers Squibb, GSK, Fosun Pharma, National Heart Centre Singapore, Mayo Clinic, Cleveland Clinic, Stanford University Institute for Human-Centered Artificial Intelligence (HAI), Stanford University, Johns Hopkins University, USC, National University of Singapore, Kameda Medical Center (Japan), Karlsruhe Institute of Technology (Germany), Université Lorraine?(France), Tecnológico de Monterrey?(Mexico), Hub de IA del Tec de Monterrey?(Mexico).
How AI Can Help Address The Global Shortage of Radiologists
There's a global shortage of radiologists. Today, over 2/3 of the people on earth do not have access to radiologists. There are big disparities between counties and within countries. AI can help radiologists triage patients and reduce delays by identifying and flagging abnormal medical images. AI can also help doctors assess which patients are are higher risk for some diseases. This article includes examples of AI /DL applications that are helping to address the radiologist shortage.
How AI Can Help Address The Global Shortage of Pathologists
There's a global shortage of pathologists. Today there are slightly more than 102,000 pathologists spread over 130+ countries worldwide. There are big disparities between regions with 2/3 of the pathologist workforce located in just 10 countries. This year,?Andrey Bychkov, MD, PhD, Director of Digital Pathology at Kameda Medical Center researched pathologists worldwide and mapped the worldwide supply. This is the first contemporary report on the global workforce of pathologists and its findings are informative. In this article I provide a list of recent milestones using AI in pathology.
Stanford Researchers Use AI To Sequence A Patient's DNA In 5 hours
In an exciting new milestone, Dr. Euan Ashley and his colleagues at Stanford School of Medicine used AI to sequence the genome of a hospitalized patient in 5 hours and 2 minutes. These researchers also hold the record for the fastest genetic diagnosis in a critical care setting - 7 hours 18 minutes! One of the patients in the study was a 3-month-old full-term infant who was having epileptic seizures. MRI imaging of the infant's brain revealed no abnormalities. The researchers used the new method and in 8 hours and 25 minutes they identified a likely pathogenic heterozygous variant in?CSNK2B. Since they diagnosed the disorder in 8 hours, the baby did not have to undergo additional diagnostic testing and the doctors were able to begin treatment right away.
Using AI To Predict Atrial Fibrillation Earlier and More Accurately
Axel Loewe?PhD and colleagues at the?Institute of Biomedical Engineering at Karlsruhe Institute of Technology in Germany?are developing new ways to predict cardiovascular diseases earlier and more accurately. Dr. Loewe leads an interdisciplinary team that is developing computer models of the human heart. They use digital twins to optimize diagnostic approaches and personalize therapies. Dr. Loewe and his team recently collaborated with scientists at universities in Italy, Spain, Brazil, and the UK in an important study using machine learning to analyze ECG signals. This study is significant because it is the first study that used only simulated data to train a machine learning classifier that then showed very good performance on real-world clinical data.
Scientists Integrate AI Tools To Improve Diagnosis of Heart Disease
In May 2022 a team of leading physicians, AI scientists, and tech experts in Singapore at the?Agency for Science, Technology and Research (A*STAR),?National Heart Centre Singapore,?National University Health System,?and?Tan Tock Seng Hospital,?announced the development of a new AI platform called?APOLLO?to help doctors improve diagnosis and treatment of cardiovascular disease. The APOLLO platform integrates several AI toolkits that analyze CTCA scans into clinical workflow. By integrating AI toolkits with diagnostic and prognostic capabilities, the platform enables them to work functionally together to provide automated anonymization, reporting, calcium scoring, epicardial fat, coronary stenosis and plaque quantification.
Insilico Medicine Announces Novel Preclinical Candidate To Treat COVID19
In May 2022 Insilico Medicine announced the nomination of a novel preclinical drug candidate to potentially treat COVID-19. The new candidate was designed by Chemistry42, Insilicio Medicine's propriety AI-driven small molecule generation platform. It is intended to be used to treat people who have been infected with SARS-CoV-2 and other coronaviruses. The drug candidate works by preventing the coronavirus from replicating. In studies it showed broad-spectrum antiviral activities, not only for SARS-CoV-2 and its variants, but also for other types of coronaviruses that cause diseases including severe acute respiratory syndrome (SARS) and?Middle East?respiratory syndrome (MERS).
Major Milestone Reached: First AI-Developed Drug Enters Phase 1 Trial
In February 2022 Insilico Medicine achieved a major milestone in the history of AI for Drug Discovery. The company announced the successful completion of their Phase 0 human study and commencement?of their Phase 1 human clinical trial with their first-in-class anti-fibrotic drug candidate. This is significant because this is the world’s first Phase 1 clinical trial of a drug developed from scratch using AI. This sets a historical precedent because Insilico Medicine used an AI-discovered molecule based on an AI-discovered novel target and did this rapidly and at a low cost. Although other AI companies are conducting Phase 1 clinical trials, Insilico Medicine's trial is the first trial in which human beings will be dosed with an AI-discovered novel molecule based on an AI-discovered novel target.
FDA Clears World's First AI-Driven Portable 3D Breast Ultrasound Scanner
In May 2022 iSono Health announced FDA clearance of the company’s ATUSA? System for breast imaging. This is world's first AI-driven portable and automated 3D breast ultrasound scanner. In just 2 minutes, the ATUSA system automatically scans the entire breast volume, independent of operator expertise, and offers 3D visualization of the breast tissue. The ATUSA system is aimed at doctors offices and women’s health clinics to allow for fast and convenient whole breast imaging at the point of care.?The system is fully automated, easy to operate and is designed to be used by physicians, nurses, and medical assistants.
Indica Labs Receives CE-IVD Mark For AI-Based Prostate Cancer Detection
In May?2022 Indica Labs?announced CE-IVD Mark certification for HALO Prostate AI. This deep learning-based screening tool assists pathologists in identifying and grading prostate cancer in core needle biopsies. Developed in collaboration with Dr. Yuri Tolkach and colleagues?from the Institute of Pathology of the University of Cologne, the platform provides a fully validated and automated end-to-end workflow in tissue-based research, clinical trials, and diagnostics. Halo prostate AI was trained using digital scans obtained from multiple scanning platforms and was designed to be scanner agnostic.?
FDA Clears AI That Helps Hospitals Reduce Missed Fractures By 29%?
In March 2022 the FDA cleared?Gleamer BoneView? AI software to detect fractures and traumatic skeletal injuries in the US. This is the first solution to receive FDA and CE clearance in this category. In a recent study BoneView?helped hospitals reduce missed fractures by 29% and increased sensitivity by 30% for exams with more than one fracture. BoneView helps detect and localize fractures over the entire appendicular skeleton, rib cage, thoracic and lumbar spine, improving sensitivity and specificity, while reducing reading time.?Boneview highlights regions of interest with bounding boxes around areas where fractures are suspected so radiologists can prioritize reading those X-rays.?
Aidoc Receives FDA Clearance For AI Algorithm To Detect Brain Aneurysm
In April 2022?Aidoc?received its 9th FDA 510(k) clearance. This FDA-clearance was granted for Aidoc's AI algorithm for triage and notification of brain aneurysm. Aidoc's AI based decision support software analyzes medical imaging data, flags findings, and alerts radiologists, surgeons, and neurologists of suspected positive cases. In 2018 Aidoc was granted FDA clearance for an algorithm to flag acute intracranial hemorrhage in CT?scans, and in 2019 the company received FDA clearance for its?pulmonary embolism?system. Aidoc has also obtained?FDA clearance for algorithms to detect pneumothorax,?stroke, cervical fracture,?intra-abdominal free gas, and?incidental pulmonary embolism.?
AI Accurately Predicts If and When a Person Will Experience Cardiac Arrest
Sudden cardiac death caused by arrhythmia is responsible for 20% of all deaths worldwide. This type of heart attack is caused by an electrical issue that stops the heart from beating correctly. Natalia Trayanova, PhD, MS and her colleagues in the?Trayanova Lab?at Johns Hopkins University are conducting cutting edge research in computational cardiology to predict this type of cardiac event. They are using machine learning to find features in medical images that reveal the condition of a person's heart and to determine risk of cardiac arrest. They were able to accurately assess a person's risk of sudden cardiac arrest for up to 10 years with high accuracy. The algorithm's predictions were significantly more accurate on every measure than doctors' predictions.
AI Identifies Hard To Detect Endoscopic Kidney Stones With High Accuracy
Scientists at?Université Lorraine?in France and?Tecnológico de Monterrey?in Mexico, in collaboration with the?Hub de IA del Tec de Monterrey?in Mexico, published a paper demonstrating the potential of machine learning and deep learning to help urologists recognize hard to detect endoscopic kidney stones in images. They compared the performance of machine learning and deep learning architectures in kidney stone recognition using incremental ML and explainable AI. They achieved promising results using deep learning methods to identify several types of hard to detect kidney stones.
AI Accurately Diagnoses Early-Stage Parkinson's Disease For The First Time
In April PreciseDx?announced that its AI based pathology technology can accurately diagnose early-stage Parkinson's disease in living patients. PD is very difficult to diagnose and postmortem assessment has been the gold standard for definitive diagnosis.?In a new study, PreciseDx’s AI algorithms were able to detect Parkinson's pathology in image patches from biopsy samples with 99% sensitivity and 99% specificity compared to expert annotated ground truth and surpassed human pathologists in accuracy. Their groundbreaking study was a collaboration with The Michael J. Fox Foundation.
Insilico Medicine's AI Identifies Novel Targets To Treat Age Related Diseases
In March 2022 Insilico Medicine published a paper demonstrating the power of AI to identify novel targets for a wide range of age related diseases. The paper details how Insilico's PandaOmics AI platform was able to predict molecular targets for new drugs to treat diseases such as Alzheimer’s, Parkinson’s, and rheumatoid arthritis. The scientists used Insilico's PandaOmics platform to search through public repositories of transcriptomic datasets. The AI retrieved and processed data and then generated a list with the top 100 molecular targets associated with 14 age-associated diseases. Another list was generated with the top 100 targets for 19 diseases not linked to aging. The AI then narrowed down the list to specific genes associated with age related diseases, and the targets were classified by their relationship to specific hallmarks of aging.
领英推荐
Fundraising, Partnerships, and Ecosystems
Aidoc and Gleamer Partner To Expand the Use of AI in Medical Imaging
In May 2022 Aidoc?announced an agreement with?Gleamer?to integrate Gleamer’s BoneView solution for X-ray in the Aidoc platform. This partnership will help health systems address the increasing volume of medical images and the worldwide?radiologist labor shortage. Integration of Boneview into Aidoc’s AI platform will give many more clinicians access to a tool to help them identify fractures in limbs, pelvis, thoracic and lumbar spine, and rib cage. Aidoc’s end-to-end AI platform already includes numerous third-party AI vendors including Imbio, Riverain, Subtle, Icometrix and ScreenPoint.
Ibex & Hartford HealthCare Collaborate To Improve Breast Cancer Diagnosis
In May 2022?Ibex Medical Analytics, a pioneer in AI-powered cancer diagnostics, and?Hartford HealthCare?launched a clinical research and innovation initiative to help physicians deliver high-quality diagnosis and improved care for breast cancer patients. Hartford HealthCare has more than 400 locations serving more than 17,000 people daily. Ibex helps diagnose cancer (including rare sub-types), grade different malignancies and detect more than a hundred other clinically relevant diagnostic features in multiple tissue types.
Deepcell Raises $73 Million to Advance AI Powered Single Cell Analysis
In March Deepcell?announced the close of a $73 million B round. Their total funding to date is $100 million. The company plans to use this funding to advance its AI powered single cell analysis platform. Deepcell combines state-of-the-art machine learning, microfluidics, imaging, and genomics to power phenotype activated cell analysis and sorting. Since Deepcell’s AI-powered technology can isolate and capture any cell population based on morphological features, it can be used for a wide variety of research, translational applications, diagnostic testing, and therapeutics.
National Heart Centre Singapore Launches AI Lab for Cardiac Imaging
In March 2022 National Heart Centre Singapore launched the?Cardiovascular Systems Imaging and Artificial Intelligence Lab. This is the first AI-driven research laboratory for multi-modal and multi-disease cardiac imaging established in Singapore and South East Asia. The CVS AI lab will drive initiatives to improve the precision and efficiency in predicting and identifying coronary artery disease, congenital heart disease, and heart failure. AI can analyze MRI and CT images of the heart and blood vessels in much more detail than the human eye and identify subtle changes. AI can provide an enhanced level of detail giving clinicians important insight of the patient’s condition.
Mayo Clinic Platform Launches Accelerator For AI HealthTech Startups
There are currently more than 200?AI?projects at Mayo Clinic. The Mayo Clinic Platform collaborates with pharma companies, medical device companies, and health tech startups to drive innovation?in diagnosis and treatment. This year the Mayo Clinic Platform announced their new accelerator program for AI powered health tech startups. The 20 week immersive?Mayo Clinic Platform Accelerate Program?is designed to help innovative startups establish credibility and get market ready so that they can spark innovation in healthcare. Startups in the accelerator will receive guidance on clinical, regulatory, technology, and business decisions. There are also opportunities for workflow integration and clinical collaboration for research trials, access thought leadership, and peer-reviewed publications.
PathAI and Cleveland Clinic Announce 5 Year Strategic Collaboration
In March 2022 PathAI and Cleveland Clinicannounced a 5 year strategic collaboration. The partnership will combine PathAI’s AI-based platforms with Cleveland Clinic’s clinical expertise and multi-modal data to unlock a broad implementation of next-generation pathology diagnostics. The objective is to accelerate innovation in precision pathology and to bridge communities in the healthcare ecosystem including patients, biopharma, and academic research. Cleveland Clinic will leverage PathAI’s quantitative pathology algorithms to conduct new translational research and for use as clinical diagnostics in multiple disease areas.
BenevolentAI & Helix Group at Stanford University Announce AI Collaboration
In February 2022 BenevolentAI?and?Helix Group?at Stanford University announced an AI research collaboration. Both BenevolentAI and Stanford have a strong track record in using AI to tackle some of the biggest challenges in biology and medicine. Through this partnership, researchers from the two organizations have pooled their expertise in AI, natural language processing and creating knowledge graphs to reason about drug discovery, with the aim of finding more effective methods to extract knowledge from biological and clinical information. The partnership aims to discover more effective methods to extract knowledge from biological and clinical information and extend the potential of AI to help scientists discover and develop better medicines.
USC Launches Center for Neuronal Longevity
A multi-disciplinary team of world-renowned researchers have teamed up to lead the new?Center for Neuronal Longevity?at USC. The new center brings together researchers to develop novel therapies to address vision loss and neurodegenerative diseases of the eye and brain.?The CNL is developing close ties with industry leaders to help accelerate the translation of groundbreaking technologies into the clinic. Twenty/Twenty Therapeutics exemplifies such a key industrial partnership. The company (a joint venture between Santen Pharmaceutical?and Verily) harnesses AI, microelectronics, and scalable digital technologies to treat eye diseases. While their current focus is the eye, their long-term goal is to apply the same technologies to alleviate neurodegenerative disorders of the brain, such as Alzheimer’s disease.
Roche - Bristol Myers Partnership Will Leverage PathAI's Platform
In March 2022 Roche announced a new collaboration with Bristol Myers Squibb that will use cutting edge digital methods to improve interpretation of tissue-based assays. As part of the collaboration, Roche will integrate a novel algorithm developed by PathAI into its digital pathology workflow software. The AI-powered algorithm will be used to analyze clinical trial samples and generate biomarker data.?AI will enable very precise diagnoses that will lead to personalized treatment options for patients with solid tumors.?Roche is leveraging its partnership with PathAI (established in October 2021) in which Roche distributes PathAI's AI-powered technology through Roche's uPath enterprise cloud software.
PathAI and GSK Partner To Accelerate Drug Development Using AI
In April 2022 PathAI and?GSK announced a strategic multi-year partnership to accelerate scientific research and drug development programs. The collaboration will focus on oncology and non-alcoholic steatohepatitis (NASH) by leveraging PathAI’s technologies in digital pathology including the use of PathAI’s AIM-NASH tool. By partnering with biopharma and clinical stakeholders PathAI leverages the power of proprietary AI and digital pathology tools to accelerate and improve precision medicine. The company's technology fuels biopharma initiatives from translational research through in vitro diagnostic device development to deliver faster, more accurate diagnoses and better personalized treatment plans for each patient.
Insilico Medicine and Fosun Pharma Form Strategic Collaboration
In January 2022 Insilico Medicine and Fosun Pharma announced an important strategic collaboration agreement. The collaboration includes an upfront payment of $13 million plus an equity investment in Insilico Medicine of an undisclosed amount. The collaboration will include an AI-driven drug discovery research and development collaboration on four biological targets, as well as the co-development of Insilico's QPCTL program. Insilico will receive an upfront payment for the R&D collaboration projects and the co-development of the QPCTL program, potential milestone-based payments, and share commercialization profits from the QPCTL program.?
How 10 AI Experts are Shaping the Artificial Intelligence Ecosystem in Italy
Ten AI experts from various sectors and with diverse backgrounds are working together to shape the Italian AI ecosystem. They are building an ecosystem that is useful, productive, and that protects the dignity of human beings. The Italian approach to AI is in line with what the EU is already doing. In fact, some of these experts are actively participating in the EU strategy-building effort. The experts assert that AI must be?human-centric?and?responsible and the ecosystem must be multidisciplinary and inclusive. Now is the time for full engagement in governance because today's discussions will influence how we experience AI tomorrow.
Highlights From The Stanford Institute for Human-Centered AI 2022 Report
The 5th edition of the Artificial Intelligence Index Report entitled?Measuring Trends in Artificial Intelligence?produced by the Stanford University Institute for Human-Centered Artificial Intelligence (HAI), is one of the most credible sources for data about AI worldwide. This year's report is 299 pages long and includes 173 visualizations. The report also includes an interactive visualization tool that can be used to compare 29 countries using 23 indicators. I selected 10 highlights to share in the DeepTech Newsletter.
This is the Spanish Edition of this Report
Inteligencia Artificial En El Cuidado de La Salud Puntos Destacados e Hitos
Este es mi nuevo informe Inteligencia Artificial En El Cuidado de?La Salud Puntos Destacados e Hitos. En este informe, destaco las implementaciones exitosas de IA, incluida la IA para secuenciar el ADN para diagnosticar enfermedades raras, la IA para predecir la fibrilación auricular, la IA para mejorar el diagnóstico de enfermedades cardíacas, la IA para descubrir un nuevo candidato a fármaco para tratar el COVID-19, la IA que predice si una persona experimentará un paro cardíaco, IA que identifica cálculos renales endoscópicos difíciles de detectar e IA que diagnostica con precisión la enfermedad de Parkinson en etapa temprana por primera vez.
This is my previous AI Milestones report published in January 2022
AI In Healthcare Highlights & Milestones Q1 2022
AI in Healthcare companies are making significant progress. There's been significant progress in AI for drug discovery, medical imaging, diagnostics, pathology, and clinical trials. Important peer reviewed papers have been published and dozens of partnerships have been formed. Big Pharma companies and major tech companies have become active in the space. I expect several technologies to converge across the healthcare spectrum this year.
Subscribe and Comment
I'm interested in your feedback - please leave your comments. To subscribe please click subscribe at the top of this article.
Copyright ? 2022 Margaretta Colangelo. All Rights Reserved. This report is the property of Margaretta Colangelo. If you would like to share the report you may share a link to my original report on LinkedIn. You may not reproduce or republish this report or any content from my report in any other newsletters, publications, or on other websites.
This article was written by?Margaretta Colangelo. ?Margaretta is Co-founder of Jthereum. She serves on the advisory board of the AI Precision Health Institute at the University of Hawai?i?Cancer Center. She's based in San Francisco?
Twitter?@realmargaretta
Senior Research Scientist at NVIDIA Research
2 年This is a very comprehensive analysis. Great job ! hope to hear more about advancements of AI in healthcare and wide adoption.
SVP - Center for Health Technology & Innovation
2 年Great report and insights here... thank your for focusing on this topic and sharing!
P&L Commercial Leader I Keynote Speaker I Author I E-pharmacy I E-health I Digital Health & IoT I Biosimilars I Biotechnology I Branded Formulations I Medical Technology I Healthcare Startups | Biker-Guitarist-Footballer
2 年Thanks Margaretta Colangelo, this is amazing as usual and super comprehensive! Its like a living encyclopedia of AI in Healthcare!
Chief Executive Officer at ONL Therapeutics
2 年Great overview Margaretta. Thanks for taking the time to share this thorough update so generously.
P&L Commercial Leader I Keynote Speaker I Author I E-pharmacy I E-health I Digital Health & IoT I Biosimilars I Biotechnology I Branded Formulations I Medical Technology I Healthcare Startups | Biker-Guitarist-Footballer
2 年This is consistently the most comprehensive post on the most meaningful & practical real life applications of AI/ML in healthcare. Thank you Margaretta Colangelo for the superlative effort of bringing it all together. Always make it a point to study all your updates in detail