Clinithink looks at the value of unstructured clinical text

One of the biggest challenges facing the digitization of healthcare is the amount of health information and insight contained in the free text of clinical notes. Clinithink’s unique technology aims to extract this information and integrate it into the comprehensive record to help fill the gaps in clinical care.

“Structured apps work well on phones, but in medicine, it’s all about the stories,” Chris Takaberry, co-founder and CEO of Clinithink, told Digital Health.

“Our solution exists because traditional technologies can do nothing with that data and yet there are millions of documents, radiology summaries and discharge notes that are inaccessible to traditional technology. We can extract meaning from that content and use it to create value. can enable.

Clinithink’s CLiX artificial intelligence (AI) enabled technology focuses on the realm of unstructured data and free text “narratives” that clinicians type into computer notes every day and by identifying patients in need of additional care and hospital admissions. Avoidance has the potential to save clinical hours and money. ,

In May, Cleanthink announced a partnership with access group To provide RIO, an electronic patient record (EPR) system for mental health, children and community health systems, covering over 30 sites in the UK.

The two companies are already working together to help mental health provider, East London NHS Foundation Trust (ELFT), identify vulnerable patients and ultimately help control costs.

extracting information from free text

The main focus of CliniThink is the progress note or episode of care report summarized in discharge summaries or clinic letters, UK trained anesthesiologists and also have a Master’s degree in Computer Science.

Its technology has been exposed to nearly 12 million medical conditions, including both common and less frequently seen ones. He notes that there are 72 different ways to say a fracture and several ways to interpret an observation whether a patient has evidence of breathlessness.

The CLiX technology can interpret text “in the presence of uncertainty, misspellings and acronyms,” says Tackaberry, and the software has been adapted for use in Spanish and Portuguese, as well as English.

Because the system only requires medical notes, the documents are generally easy to retrieve from the EPR and other databases that health systems typically use, Takaberry says, adding that the software is “completely EPR agnostic.” ” Is. The company operates in a niche area with a few other vendors operating in the space.

As a result of the success at ELFT, the technology is being introduced in a number of other health trusts. Over the past two years, Barts NHS Trust in London has been using Clinithink solutions to identify people at risk of worsening diabetic foot disease (DFD).

“The general principle is very exciting in the context of population health because, taking the example of the diabetic foot, the information that tells vascular surgeons and podiatrists about people at risk is not available in structured data, it is in narrative form,” Takaberry said. They say.

The company is working on other specific applications for trusts in the Midlands, Herefordshire and Worcestershire.

Clinithink is now looking to move beyond identifying patients who meet certain criteria only as described in the free text to finding patients who are more broadly high risk or vulnerable. I

In addition to preventing bad outcomes and unnecessary clinical spending, the company plans to work with drug companies to identify eligible people for clinical trials.

“Using traditional machine learning to make predictions in healthcare is notoriously difficult compared to other industries, but information that might indicate someone might be getting worse is out there in the free text,” Takaberry says.

“If you take information from our[technology]and put it into a machine learning stack, you can support finding a needle in a haystack for population health. ,