Peer-reviewed articles show clear benefits of machine learning for healthcare applications

Cambridge, Mass – February 8, 2017 – Cyft Inc., a leading provider of machine learning solutions for healthcare, is the result of a decade of research into how machine learning and natural language processing technology can improve healthcare. Cyft is pleased to recognize the publication of three peer-reviewed studies by leading health services researches that used the research precursor to Cyft to address important clinical questions.

The first study appeared in the Journal of Surgical Research and was led by Dr. Caprice Greenberg of the University of Wisconsin School of Medicine and Public Health. The study demonstrated that an early research version of Cyft could process millions of structured and unstructured data points from electronic medical records that would otherwise only have been accessible via manual chart review. As a result, the team discovered differences in the surgical treatment of young women with breast cancer at 58 different community hospitals.

The second study involving Cyft in 2016 was led by Dr. Caitlin Brennan of the NIH Clinical Center. Appearing in the Journal of Nursing Management, “Feasibility of Automating Patient Acuity Measurement Using a Machine Learning Algorithm” showed that unstructured and unstructured clinical data can be used to automatically classify the acuity of patients in an inpatient setting, allowing nurse managers to more appropriately allocate personnel.  

The third study appeared in the Journal of Patient Safety and was led by Dr. Brian Shiner of Dartmouth’s Geisel School of Medicine. In it, a team of health services researchers automatically detected and classified falls occurring in a hospital using the free text of electronic health records.  Despite a small sample, they outperformed previous research efforts on the same task by a significant margin.

Cyft founder and CEO Dr. Leonard D’Avolio considers participation in research critical to the company’s progress and mission: “We gain a tremendous amount from working with world class researchers on difficult problems. It pushes our understanding and ability while generating important clinical evidence that can be used to improve care.”