Machine Learning Model Accelerates Healthcare Record Processing by 87

Maruti Techlabs' client, UKHealth, is one of the largest healthcare service providers in the UK, managing thousands of hospitals and care units, including multispeciality hospitals, diagnostic clinics, and pharmacies. They aim to make healthcare services more efficient and accessible. Disclaimer - The name UKHealth is a placeholder as there is an NDA signed between Maruti Techlabs and the client.

Maruti Techlabs' client, UKHealth, oversees the data management of thousands of hospitals and clinics across the UK. Doctors at these facilities generate hundreds of thousands of discharge, referral, and follow-up letters daily. UKHealth's data teams manually assess and categorize these letters into specific categories, such as discharge, follow-ups, and referrals, for their central Hospital Information Management System (HIMS). This manual process is time-consuming, resource-intensive, and prone to errors and discrepancies.

Maruti Techlabs designed a machine learning model for UKHealth to automatically extract and classify data from diagnosis letters into discharge, follow-ups, and referrals categories. The process involved two steps: text extraction via OCR and phrase identification via NLP. First, physical diagnosis letters were scanned and converted into a structured dataset using OCR, which processed the text for the model. Then, an NLP algorithm interpreted the text to classify the letters based on context-specific phrases. The model was integrated with UKHealth's Hospital Information Management System (HIMS) for automatic updates and efficient patient record management. Initially, the model had a confidence score of 80%, which improved to 85% after six months of supervised training.