Responsible AI
summary
Describes building applied medical AI that actually ships: a breast‑cancer pipeline for Afghan clinics using heavy preprocessing, augmentation and custom models developed over 6–8 months to address tens of thousands of late‑stage cases; early experiments with capsule networks lead to pragmatic adoption of convolutional nets; an ophthalmology appliance trained on 70,000 images reaches high accuracy and gains regulatory approval; dental radiograph prototypes use Mask R‑CNN and generative/flow models. Results work but need more data and tighter recall metrics.
AI-written summary · about
chapters (7)
- 0:00Start
- 1:43INTRO
- 3:21Our Management Team
- 6:35What companies need to do to become an Al Company
- 16:35Story #1: Healthcare Client fights breast cancer with Al
- 21:47Fighting breast cancer with Al
- 24:54Story N2: Diabetic Retinopathy in China
related talks
37mAI in medicine: where do we stand?
26mFrom Pixel to Patient: Making Medical AI Work in the Real World
06mTransitioning AI from the lab to real-life applications
18mMedical AI Beyond Prediction: From Normative Learning to Agentic Healthcare AI
09mHow AI is changing the healthcare system
47mAI for making decisions in clinical medicine
24mDeveloping Innovative Systems Through Responsible AI In Healthcare
46mExplainable AI for Biomedical Image Processing