Improving the quality care and identifying errors
The Story
As the digitization of healthcare continues, the industry is now taking the opportunity to scale up their big data defenses and develop the cutting-edge infrastructure required to meet the imminent challenges. The increasingly growing number of applications of machine learning in healthcare allows organisations to glimpse at a future where data, analysis, and innovation work hand-in-hand to help countless patients without them ever realizing it. Our client is Lifesalus, an oncology company in the UK involved in cancer treatments.
The Challenge
The major challenge for our client was spending too much time on patient documentation and clinical decisions. Due to physician burnout, patients have suffered from misdiagnosis or their real problems have gone undetected and left to worsen. Lifesalus asked us for a cutting-edge solution to improve the daily workflow:
- administrative tasks as well as billing information answering medical questions through standardized processes
- a predictive model to identify patients who need periodic care
Artificial intelligence in healthcare is able to change the process of physician assessment and patient diagnosis, reducing the time and human effort needed in carrying out routine tasks.
We sourced a team composed by Machine Learning engineers, one project manager, Data Scientists who have proposed a solution based on the Natural Language Processing (NLP) to enhance the client’s productivity. NLP has the great potential to identify errors in care delivery.
The project manager oversaw the project and kept track of the progress making sure that the project was delivered at the defined deadline.
Instead our Data engineers created data pipelines through the integration between the new framework and the existing Lifesalus ICT infrastructure. Our Machine Learning engineers delivered project’s milestones:
- Understanding human speech and extracting meaning for testing and training the virtual machines.
- Unlocking unstructured data in documents and databases by abstracting out key concepts and values and making this information available for decision support and analytics.
The project lasted 3 months where our data experts cooperated with the in-house team for improving the daily workflow of the client’s staff.
The Outcomes
Artecha’s team developed a powerful AI solution which supports the client’s staff from the diagnosis to administrative tasks. The intervention areas were:
- Clinical Trial Matching.We combined NLP and ML to process unstructured electronic health record (EHR) documents, patient data pathology reports, operating notes and other important medical data in free-text form that cannot be searched easily.
- Clinical decision support. We used the deep learning approach to integrate the Clinical Decision Support (CDS) which offers information, insights and expertise to aid physicians, staff, and patients.
Main Features
- An useful AI platform for physicians and patients. The AI platform relies on Free-form spoken or text query useful for queries that require gathering and organizing data from multiple source
- An algorithm classifies patients who meet the core eligibility criteria of the clinical trial in order to dramatically minimize the pool of potential personnel screening candidates.
- Computer assisted coding (CAC) framework, powered by NLP algorithm connected the documentation in EHR with transcription systems and financial systems in the healthcare field. With NLP, the CAC web-based application looks for specific phrases and associates them with the right medical codes directly from clinical documents.
Team: Machine Learning engineers, Data scientists, Data engineers, one Project Manager