Per WSJ, Johnson & Johnson is making a big bet on using data science (DS) and artificial intelligence (AI) to support its healthcare work. They have hired 6,000 data scientists/digital specialists in recent years and spent hundreds of millions of dollars on their work, such as using machines to evaluate massive health-record data sets. Last year, they opened an advanced DS center near San Francisco.
Initial efforts focused on diagnostics, such as an algorithm that analyzes heart tests to spot a deadly type of high blood pressure much sooner than humans can, and voice recognition technology to analyze speech for early signs of Alzheimer’s. The long-term goal is using DS/AI for drug discovery. While some are skeptical, J&J says it has an edge due to a massive database called med.AI that has real-world, anonymous data collected from patient visits to doctors/hospitals, also years of clinical trial test results. This process helped them develop an experimental cancer drug scheduled to start human testing in 2024.
J&J is now adding DS into most of their drug development projects. They are also increasing collaborations, including >50 partnerships with DS startups and others. In one recent project, J&J led a group of 13 drug companies that analyzed blood samples from 50,000 people in the UK as part of a national database called UK Biobank. They identified thousands of genetic variants that influence levels of certain blood proteins, about 80% of which weren’t known previously. J&J plans to analyze the data set using AI/machine learning to help spot patterns, which could lead to new drugs/diagnostics that target the gene-protein links to various diseases.
J&J is also using AI to study digitized images of biopsies to detect differences between tumors, which could help identify genetic subtypes for certain tumors. They could then use this information to make medicines that target specific genetic sub-types.
Another project with Mayo Clinic and Anumana, a health technology company, aims to speed up diagnosis of high blood pressure in the lungs (pulmonary hypertension – PH). It can now take >2 years to identify the disease. J&J and its partners gathered 6 million anonymous patient records, including more than 8 million electrocardiogram readouts, then fed the records into a software algorithm to teach it to spot patterns in the electrical readings. Using the algorithm in conjunction with ECGs could cut the time to a PH diagnosis by 12-18 months.
MOSIMTEC specializes in simulation modeling and digital twins, driven by data science in everything we do. We have worked with many large pharma companies on a diverse range of projects, typically focusing about 40% of time on data-related issues to ensure needed data quality/integrity. We work with big data every day and know how to make it work for you. MOSIMTEC data scientists can help future-proof your business.
#pharma #drugdevelopment #datascientists #artificialintelligence #AI #digital twin in healthcare
Read More: https://www.wsj.com/tech/biotech/johnson-johnson-hiring-data-scientists-ai-ccbf2c07