Healthcare AI Daily Briefing

Diagnostics & Pathology

Deep learning framework predicts rectal cancer pCR with internal AUC 0.92 and external AUC 0.85‑0.86.

Journal article

In 538 locally advanced rectal cancer patients from five hospitals, a deep learning framework using pretreatment 3D transrectal ultrasound yielded an internal validation AUC of 0.92 (accuracy 86.2‑90.8%) and an external validation AUC of 0.85‑0.86 (accuracy 81.6‑83.5%); decision‑curve analysis showed clinically meaningful net benefits. [1]

Research & Evidence

French oncology pharmacists widely use AI but mostly without validated tools.

Journal article

A national survey of 56 responding French oncology pharmacists found 94.7% reported AI adoption, yet 65.3% operated within a shadow AI framework and only 17.3% had access to fit‑for‑purpose validated tools. Primary expectation was time savings on repetitive tasks (83.9%), while key barriers included fear of erroneous outputs (64.3%) and patient data confidentiality concerns (62.5%). [2]

Industry & Products

ARPA-H awards contracts to six teams to build agentic AI for heart failure.

ARPA-H awarded contracts to six teams, including Kaiser Permanente, Duke University, Stanford University and AI vendors Atman Health, Tempus AI and Updoc, to build an FDA-authorized autonomous AI agent for heart failure patients. The four-year contract totals $62.7 million, with $33.7 million planned for the first year. The FDA has not yet approved an agentic AI tool for clinical care. [3]

Clinical Practice

AI‑enhanced EHRs reduce clinician burnout by 13% in multi‑system study.

A study of 263 physicians and advanced practice practitioners across six healthcare systems reported that after 30 days of using AI‑integrated EHRs, burnout decreased from 51.9% to 38.8%, a 13‑point reduction. The AI assistance surfaces information, reduces repetitive tasks, and supports workflow actions, aiming to restore the human connection in care. [4]

Operations & Workflow

AI scribes may increase documentation codes and costs.

Podcast from Lifers (Second Opinion)

A speaker reported that automating junk tasks with AI scribes could contribute to higher documentation codes and add expense. [5]

Industry & Products

GenHealth.ai raises $16.5 million Series A to automate healthcare admin tasks.

GenHealth.ai announced a $16.5 million Series A round led by Flare Capital Partners, with participation from Craft, Obvious Ventures, Eniac Ventures and others, bringing total funding to $30 million. The company says its AI agents for intake, eligibility, prior authorization, billing and appeals can increase revenue by 34% and cut administrative costs up to 80%. [6]