Usecases
Crisis Text Line identifies 85% of high-risk users …
The Crisis Text Line allows users in need of immediate mental health help to text a counsellor. Some of those users need emergency service intervention, so machine learning has been applied to discover relations between text message content and the requirement for immediate intervention, allowing the service to better prioritize users. The system identifies 86% of users at high risk …
NCI designated cancer centers reduces time to treatment …
NCI-Designated Cancer Centers have implemented Digital Reasoning’s deep analytics care management software to automate processing pathology and radiology reports. The software helps speed up patient's journey from diagnosis to treatment by discovering, classifying and prioritizing cancer cases for optimal follow-up. It augments health care providers by supporting the workflows, triage and extracting key points from documents.
Amazon announces its Comprehend Medical platform to analyse …
Amazon has announced a machine learning service for analyzing and extracting information from different unstructured medical records and text. Comprehend Medical is accessible for developers through AWS, and is partnering with healthcare research centers to assist with things like clinical trial patient matching.
The New York Times automates historic photo content …
The New York Times is digitizing their repository of several million historic photos using Google Cloud Services. Machine learning is used to identify and extract bits of text in the photos automatically, adding important content and context to the files without human intervention.
MARS Marketing improved accuracy of purchase intent in …
MARS Marketing conducted a marketing research study which tracked customers' facial emotions while watching ads. This was to determine if emotional state could be used to predict likelihood to buy products, particularly more so than traditional marketing research techniques. They found that emotional state complemented research subjects' self-reported purchase intents to boost predictive power.
JPMorgan reduced lawyers' hours by 360,000 annually by …
JPMorgan uses its proprietary machine learning software (COIN - Contract Intelligence) to automate the work usually undertaken by lawyers to analyse commercial loan agreements by extracting contract attributes in a few seconds.