Code for good
- 1. Renee Yao, Global Healthcare AI Startups Lead, NVIDIA | Sep 2019 CODE FOR GOOD
- 3. 3 NVIDIA IN HEALTHCARE BREAKTHROUGHS INSTRUMENTSECOSYSTEM RESEARCHIMAGING DRUG DISCOVERY
- 4. 4
- 5. 5 AI-ACCELERATED DRUG DISCOVERY SENSORS PERCEIVE REASON PLAN DATA DATA ANALYTICS MACHINE LEARNING ACTUATORSAI MODEL 11B 6 yrs $100B X0%
- 6. 6 AI DETECTS GROWTH PROBLEMS IN CHILDREN Detecting growth-related problems in children requires calculating their bone age. But it’s an antiquated process that requires radiologists to match X-rays with images in a 1950s textbook. Massachusetts General Hospital, which conducts the largest hospital-based research program in the United States, developed an automated bone-age analyzer built on the NVIDIA DGX-1 with CUDA. The system is 99% accurate and delivers test results in seconds versus days. 1950s 75yrs 1+day 2017 99% Seconds
- 7. 7 In the healthcare industry, there exists mountains of underutilized data. 16 Bit is using AI to unlock the value hidden in this data to augment diagnostics and improve the quality and accessibility of healthcare. 16 Bit is using GPU-accelerated deep learning and big data to assist radiologists in detecting breast cancer, analyze CT scans of the brain to exclude acute diseases, and accurately measure pediatric bone age. Their Bone Age analyzer has an accuracy rate of +/- 4 months and returns results in milliseconds―earning 16 Bit 1st place in the 2017 RSNA Machine Learning Challenge. 16 BIT.AI RSNA BONE AGE CHALLENGE WINNER Try the 16 Bit algorithm: 16bit.ai/bone-age
- 8. 8 AI AUGMENTS IMAGING WORKFLOWS Medical imaging analysis is a time sink for radiologists, but crucial for diagnosing many ailments. Arterys’ FDA- cleared cloud-based clinical oncology solutions use deep learning to accelerate the interpretation of CT and MRI scans. With algorithm training done on NVIDIA Tesla GPU accelerators in the Google Cloud Platform and inference occurring on Tesla GPUs running in the Amazon cloud, the automated intelligent imaging network helps clinicians better manage their busy workflows. 3 mins 10 mins 30m ppl 15-90 seconds 170 physicians 100 hospitals 30,000 cases A few radiologists BEFORE AFTER
- 9. 9 Diagnosing Retinopathy of prematurity (ROP) ―a condition that can cause blindness in preterm babies― is subjective. Doctors compare the infant’s retina with decade’s old photos to determine severity of ROP. Using a dataset of 6,000 images matched with expert ROP diagnoses and an NVIDIA DGX-1, researchers at the Martinos Center trained a deep neural network to differentiate ROP severity. Still work in progress, this method could one day help prevent blindness in infants worldwide. AUTOMATING THE DIAGNOSIS OF INFANT BLINDNESS 14k-16k 400-600 10k images <31wks <2 ¾ lbs 3.9m 28,000
- 10. 10 AI ADVANCES THE FIGHT AGAINST BREAST CANCER Breast cancer is the second leading cause of cancer death for women worldwide. Genomic tests help doctors determine a cancer’s aggressiveness so they can prescribe appropriate treatment. But testing is expensive, tissue-destructive, and takes 10-14 days. Case Western Reserve is using GPU-based deep learning with CUDA to develop an automated assessment of cancer risk at 1/20th the cost of current genomic tests. 10-14 days $4000 Auto $200 BEFORE AFTER
- 11. 11 Heart disease is the world’s biggest killer. Until recently, the best test to diagnose heart disease was an invasive and costly angiogram. HeartFlow applies GPU-accelerated deep learning to deliver non-invasive diagnostics. Trained on CT scans and computational fluid dynamics, HeartFlow creates a personalized 3D model of a patient’s coronary arteries and analyzes the impact of blockages. Heartflow reduces the need for an angiogram by 61% and healthcare system costs by 26%. HEART SMART: AI THAT DETECTS HEART DISEASE Image is not representative of actual product. Courtesy of HeartFlow, Inc. 9 million biggest 61% 26%
- 12. 12 AI ENABLED HEALTHCARE AT SCALE The human retina contains diagnostic markers for many diseases. SocialEyes hopes to transform healthcare systems worldwide with its low-cost AI enabled Mobile Autonomous Retinal Evaluation (MARVIN). MARVIN –which runs on GPU-powered Android tablets- could help tens of millions of community healthcare workers and physicians immediately diagnose a wide range of conditions for timely and effective intervention. Marginal image quality with low resource settings Improved visibility of retinal features with MARVIN MARVIN’s AI detects high-risk cases Diabetic retinopathy images courtesy of SocialEyes 415 million diabetes 50% 25 yrs 2.5 million/yr
- 13. 13 ACCELERATING TREATMENT FOR ALZHEIMER’S DISEASE Neurodegenerative diseases like Alzheimer’s affect nearly 50 million people worldwide but still lack a cure. Researchers at the Université de Reims created GPU-powered software that analyzes the effects of potential drug treatments. The interactive interface processes hundreds of petabytes of brain scan data at ~50 frames/second enabling scientists to more efficiently annotate key cells for a much faster analysis workflow. Hundreds of slices 120GB each slices 100s PB of data hrs vs. days 50m ppl 9.9 million every year 1 new case every 3.2 seconds 6th leading cause
- 14. 14
- 15. 15
- 16. 16 BUILDING A COMMUNITY Bioinformatics for Next Gen Sequencing
- 18. 18 MIXED REALITY FOR RADIOLOGISTS
- 19. 19 SOLVING THE MOST DIFFICULT PROBLEMS
- 20. 20
- 21. 21