en
0.25
0.5
0.75
1.25
1.5
1.75
2
Machine learning techniques for predicting complications and evaluating drugs efficacy
Published on May 13, 20143025 Views
In this talk we focus on potential contribution of machine learning methods to healthcare and focus on the somewhat new trend called real world evidence or post launch monitoring. We review the machin
Related categories
Chapter list
Machine learning techniques for predicting complications and evaluating drugs efficacy00:00
Outline02:45
The Medical domain03:17
Chronic diseases - prevalence and cost04:22
What is real world evidence?05:13
Prediction models are based on the data features07:19
RWE Potential value per sector10:16
Watson for Epilepsy UCB-IBM announcement10:38
Outline: Machine Learning in a nutshell11:45
Machine Learning13:01
Classification Algorithms17:21
Probabilistic graphical models19:29
What are Bayesian Networks useful for?25:22
Rules of Probability26:27
Bayesian Networks28:44
Factorization properties29:28
Outline: Three concrete examples - 145:06
EuResist Partners45:24
Predictive analytics - Therapy optimization - 150:39
Labeling therapies and prediction engines51:08
Predictive analytics - Therapy optimization - 253:06
EuResist53:17
From EMRs and blood tests to standard datum for automatic learning56:04
Comparison of performances56:31
Outline: Three concrete examples - 201:03:45
Personalized Management of Chronic Diseases01:03:49
Data Representation01:05:12
Data01:06:49
Learning Models01:08:46
Feature Selection and Dimensionality Reduction01:09:06
Do Physicians Affect Patients’ Outcome?01:10:55
Patient-Physician Match - UC Visits01:12:20
Patient-Physician Match - HbA1C01:14:52
What it is good for?01:14:54
Outline: Three concrete examples - 301:15:03
Watson for Epilepsy01:15:08
Work performed with ...01:18:56