A cardiologist with a background in medical technology, computer science and artificial intelligence has launched a product aimed at legal professionals and medical expert witnesses, targeting the tedious task of sifting through and analyzing thousands of pages of medical records.
The product, Quench SmartChart, uses generative AI to streamline the forensic review process, allowing users to quickly extract, summarize, and create timelines from large, disorganized medical record PDFs.
The product also includes AskQuench, a natural language chat feature that allows users to interact with and query records to gain key insights.
The company has just opened a waitlist for people who want early access to the product ahead of its official launch.
A lengthy process
The company's founder and CEO, Michael Loesch, is a cardiologist and adjunct professor at the University of California, San Francisco School of Medicine who left his medical practice to become a medical device entrepreneur.
After starting several successful medical device companies, he turned his attention to digital health and began studying how large-scale language models could help streamline the workflow for doctors reviewing medical records.
During his research, he found that the process for reviewing medical records by attorneys, expert witnesses and independent medical examiners in medical malpractice, personal injury and workers' compensation cases remains outdated and time-consuming.
Although records are digitized, they are often delivered as lengthy PDFs containing thousands of pages, requiring lawyers and experts to manually sift through them to understand their contents before reviewing and analyzing them – a process that can take hours or even days.
Unable to find anything on the market that could help solve this problem, Resch and his team of doctors and data scientists decided to apply the advanced AI they had already built to another product to help in this medico-legal situation.
Resh says the company developed SmartChart using its own AI engineering based on the team's expertise in healthcare. “It's not ChatGPT, it's not just a wrapper,” he says. “We have our own very sophisticated engineering.”
Using SmartChart
To use Quench SmartChart, users first upload their medical records (which can be in PDF, Word, or text format).
SmartChart uses the uploaded records to create a chronology of the case. Please note that in this invite-only phase of product development, some planned features are not yet available. Currently, the chronology appears as a list of major events in the patient's medical history, such as consultations and tests.
The timeline links to specific records detailing each event, allowing users to view the original documents in context.
Future versions of the product will also include a visual timeline, as shown in the image below, where users can zoom in on parts of the timeline to focus their review.
The chat feature, AskQuench, allows users to ask questions and gain insights about documents. In the example below, the user asks, “What is the nature of this worker's injury?” AskQuench describes the worker's injury, the cause of the injury, the treatment received, and provides an answer hyperlinked to the original document, along with a list of source documents that support the answer.
Another feature “coming soon” to Quench is the ability to generate reports based on medical records. After reviewing the timeline and asking questions using chat, users can ask Quench to generate a report. Quench will generate a detailed report covering topics such as illness or injury history, the patient's past medical history, and current symptoms.
Quench will include several templates for different types of reports, such as one for an independent medical examiner in a workers' compensation lawsuit or one for a medical expert in a medical malpractice lawsuit, and it also plans to allow users to add notes to be included in the report as they review records.
Resch also plans to make future versions able to identify missing records, allowing lawyers and experts to know what's not in the records but should be in the records — for example, if the records show an MRI was ordered for a patient, but the records don't include a report showing the MRI was performed.
“AI can help you find things that should be there but aren't and make a list of them,” he said. “This could be significant because we've heard from lawyers that there was one document that was smoking gun evidence, but somehow it wasn't included in the document.”
Another planned feature for Quench will be the ability to retrieve relevant references from the medical literature and incorporate them into the final report.
Open the waiting list
So far in developing the product, Resch has focused on recruiting physicians as users, and he wanted to make sure the product would work for them and meet their requirements.
But once the product enters its invite-only rollout phase, Quench will be ready to expand its user base to lawyers and legal professionals, particularly those working in personal injury, medical malpractice, insurance defense, class action lawsuits and workers' compensation.
At this time, there is no word yet on pricing for the product, but during the invite-only period, users will not be charged.
Resch said the best thing about SmartChart for both lawyers and doctors is that it can do something in 10 minutes that would have taken 10 hours.
“It's impossible for a doctor or lawyer to sift through 5,000 pages of documents and pick out everything they need, so it takes 10 minutes instead of 10 hours. And it's instantaneous. We don't have to send it overseas and wait a week or have a paralegal or nursing assistant look at it for days.
“This saves a lot of time compared to manual review and gives better, more accurate information.”
