ASGE Annual Postgraduate Course: Clinical Challenges and Real-Time Decision Making in GI Endoscopy (On-demand) | May 2023-How is Machine Learning different from Statistics? The thought process

How is Machine Learning different from Statistics? The thought process

Video Transcription

Video Summary

In this video, Shravanthi discusses the differences between machine learning and statistics. She begins by explaining that as clinicians, it is important to evaluate statistical technologies and understand how a machine reaches a certain inference, as we move towards using machine learning applications in clinical practice. Shravanthi highlights the differences between statistics and machine learning, emphasizing that machine learning focuses on accuracy and performance, while statistics emphasizes inference and the relationship between variables. She describes how machine learning algorithms work, starting with a messy and noisy dataset that is cleaned before the algorithm processes the data to identify patterns. She also mentions the challenges faced in developing machine learning algorithms, such as insufficient training data, data not representing real-world situations, and the risk of overfitting or underfitting data. Shravanthi emphasizes the importance of understanding and interpreting machine learning algorithms, as they can sometimes identify irrelevant features but still achieve high accuracy. She concludes by mentioning the need for explainable and transparent machine learning models in clinical practice.

Asset Subtitle

Sravanthi Parasa, MD

Keywords

machine learning

statistics

clinical practice

accuracy

inference