AI & Healthcare Newsletter #1

Bringing you the latest in AI-driven healthcare, precision medicine, and digital health innovations.

?? In This Issue:

?? What to Watch

?? FDA Issues Draft Guidance on AI in Drug Regulation

?? Read More

• The Food and Drug Administration (FDA) has released a draft guidance outlining considerations for using artificial intelligence (AI) in regulatory decision-making for drug and biological products.
• The guidance offers recommendations to sponsors and interested parties on using AI to generate data supporting regulatory decisions about the safety, effectiveness, and quality of drugs
• It introduces a risk-based credibility assessment framework for establishing and evaluating the credibility of an AI model for a specific context of use (COU).
• The framework involves a 7-step process, including defining the question of interest, defining the COU, assessing AI model risk, and developing and executing a plan to establish AI model credibility.
• The FDA emphasizes that the level of oversight, stringency of assessments, and documentation should align with the AI model's risk and the specific COU.
• The guidance highlights the importance of life cycle maintenance to ensure the AI model remains suitable for its intended use throughout the drug product life cycle.
• The FDA encourages sponsors to engage with the agency early to discuss AI model risk and appropriate credibility assessment activities.

?? In the News

1. ?? CertisAI & CertisOI Assistant personalize cancer treatment with AI-driven predictions
2. ?? FDA grants breakthrough status to an AI model improving lung cancer risk assessment
3. ?? AI chatbots match human genetic counselors, expanding access to genetic testing
4. ?? AI Predicts Regulatory Genomics ‘Language’
5. ?? AI outperforms doctors in predicting stroke-related hematoma expansion
6. ??? AI diagnoses rare retinal disease, enabling faster gene therapy treatment
7. ?? AI repurposes adalimumab for Castleman’s disease, leading to two years of remission
8. ?? AI imaging tracks pulmonary fibrosis progression, enhancing clinical trials
9. ?? Eko Health’s AI detects pulmonary hypertension using heart sound analysis
10. ?? AI-driven RNA drug discovery platform launched, backed by $9M in funding
11. ?? Apple & Brigham study AI’s role in personal and mental health management
12. ?? FDA warns smartphone settings may block diabetes device alerts
13. ?? AI matches dermatologists in diagnosing rare and complex skin diseases

?? Get the details

?? AI Transforms Cancer Research with CertisAI?

?? Read More

Certis Oncology Solutions has launched CertisAI? and CertisOI Assistant?, AI-powered platforms that improve cancer drug development.

• CertisAI? predicts drug response based on genetic markers.
• CertisOI Assistant? helps researchers analyze complex genomic data.
• These tools aim to improve personalized treatment and reduce drug trial failures.

?? Peter Ellman (CEO): “AI can finally bring true personalization to cancer therapy.”

?? FDA Approves AI Tool for NSCLC Prognosis

?? Read More

The Serial CTRS AI tool, developed by Onc.AI, has received FDA breakthrough status for non–small cell lung cancer (NSCLC) prognosis.

• The model classifies patients into high-risk and low-risk categories.
• Clinical trials showed better survival predictions than traditional CT scan assessments.
• This could enhance treatment personalization and clinical trial design.

?? Onc.AI CEO Akshay Nanduri: “We’re empowering oncologists with automated, real-time insights.”

?? Chatbot Matches Standard Genetic Testing Services

?? Read More

A study found that AI-powered chatbots are as effective as human genetic counselors for cancer genetic testing.

• 12.3% of chatbot users completed genetic testing vs. 13.6% of those who had human counseling.
• The chatbot delivered pre-test education and scheduling assistance.
• AI could increase genetic counseling accessibility while freeing up human counselors for complex cases.

?? AI Predicts Regulatory Genomics ‘Language’

?? Read More

The EpiBERT AI model, developed by Dana-Farber and Google, predicts how genes are regulated across cell types.

• AI identifies regulatory DNA sequences and chromatin accessibility.
• The model can predict gene expression changes, aiding drug development.
• Works similarly to large language models (LLMs) like ChatGPT but for genomics.

?? AI Outperforms Doctors in Predicting Stroke-Related Hematoma Expansion

?? Read More

AI models using radiomics and deep learning outperformed clinicians in predicting hematoma expansion (HE) in intracerebral hemorrhage (ICH).

• The best AI model achieved an AUC of 0.68, compared to 0.59 for expert clinicians, improving early intervention strategies.
• This advancement could help optimize stroke management, reducing complications and improving patient outcomes.

?? AI Helps Diagnose Rare Retinal Disease

?? Read More

AI identified a mutation in the RPE65 gene, enabling a patient in Korea to receive Luxturna gene therapy for retinal degeneration.

• Diagnosis took just five months, much faster than traditional methods.
• Highlights AI’s role in rare disease diagnosis and personalized treatment.

?? AI Repurposes Adalimumab for Castleman’s Disease, Leading to Two Years of Remission

?? Read More

An AI platform identified adalimumab, a TNF inhibitor, as a potential treatment for idiopathic multicentric Castleman’s disease (iMCD).

• A critically ill patient responded to the AI-recommended therapy, achieving two years of remission.
• This breakthrough highlights AI’s ability to repurpose existing drugs for rare and life-threatening diseases.

?? AI in Pulmonary Fibrosis Drug Trials

?? Read More

Qureight and Avalyn Pharma are integrating AI-driven imaging into progressive pulmonary fibrosis drug trials.

• AI will track fibrosis progression more accurately in clinical studies.
• Partnership supports inhaled pirfenidone therapy, reducing systemic side effects.

?? Eko Health’s AI Detects Pulmonary Hypertension Using Heart Sound Analysis

?? Read More

Eko Health has developed an AI-powered tool that detects pulmonary hypertension (PH) by analyzing heart sounds.

• The AI model achieved an AUROC score of 0.79, showing strong accuracy in identifying elevated pulmonary artery pressures.
• This non-invasive, low-cost screening tool could improve early PH diagnosis and treatment.

?? AI-Driven RNA Drug Discovery Platform Launched, Backed by $9M in Funding

?? Read More

Terrain Biosciences has launched an AI-powered platform to accelerate RNA-based drug and vaccine development.

• The company raised $9 million in seed funding to optimize RNA sequence design for better efficacy and manufacturability.
• AI-driven insights aim to enhance RNA stability and durability, speeding up the development of personalized medicines.

?? Apple & Brigham Study AI’s Role in Personal and Mental Health Management

?? Read More

Apple and Brigham and Women’s Hospital have launched the Apple Health Study to explore AI’s impact on health monitoring.

• The study will analyze data from iPhones, Apple Watches, and AirPods to assess links between sleep, mental health, and heart health.
• Findings could inform future AI-powered health tools, improving early detection and personalized care.

?? FDA Warns About Missed Alerts in Smartphone-Compatible Diabetes Devices

?? Read More

The FDA issued a warning about smartphone-connected continuous glucose monitors (CGMs) and insulin pumps.

• Software updates & settings changes can cause missed safety alerts.
• Patients should regularly check alert settings to avoid severe complications.

?? AI Matches Dermatologists in Diagnosing Rare and Complex Skin Diseases

?? Read More

AI models performed as accurately as board-certified dermatologists in diagnosing skin diseases.

• AI systems like ChatGPT-4o and Gemini 1.5 Pro achieved 70-80% accuracy, particularly excelling in rare cases.
• This technology could improve dermatology workflows, assisting specialists with faster and more accurate diagnoses.

?? Subscribe for weekly updates and check out LucidQuest's YouTube channel for daily insights!

?? Stay tuned for more AI-powered breakthroughs in healthcare!