Artificial Intelligence in Health Care: Liver Fibrosis, Prostate Cancer, Safer Driving

Machine learning is revolutionizing healthcare.
Here are some highlights of how artificial intelligence is changing medicine.

“People worry that computers will get too smart and take over the world, but the real problem is that they’re too stupid and they’ve already taken over the world.” 
Pedro Domingos

1) Predicting Liver Fibrosis Using Ultrasound

Liver fibrosis is caused by diseases that can damage the liver. Millions of people are affected by obesity, alcoholism, and hepatitis; All diseases that cause a vicious cycle of inflammation.

The repeated inflammation causes scarring which eventually decreases liver function and eventually requires a liver transplant.

To accurately diagnose the degree of liver fibrosis, biopsy is the current gold standard. Liver biopsy is a costly and invasive surgical procedure that has potential life threatening risks and is prone to sampling error.

Non-invasive methods of evaluating liver fibrosis are becoming more popular. In this study, researchers combined doppler ultrasound data with artificial neural networks to create a predictive model for liver fibrosis. The future of diagnosing and predicting liver fibrosis may lie in non-invasive ultrasound.

Read more here: Doppler ultrasonography combined with transient elastography improves the non-invasive assessment of fibrosis in patients with chronic liver diseases

2) Early detection of Prostate Cancer

Prostate cancer is a slow growing disease that affects millions of men each year. The earlier it is detected the more likely aggressive surgery can be avoided and the cancer can be treated with medication instead. The current gold standard for diagnosis of prostate cancer uses a blood biomarker which is not always accurate. Therefore developing a new technique for prostate cancer detection using MRI may provide a useful, more efficient means of accurately diagnosing prostate cancer. In this article, researchers describe how they used a Generalized Gauss-Markov Random Field (GGMRF) image model to classify MRI images that contained prostate cancer.

Read more here: Computer-aided diagnostic tool for early detection of prostate cancer

3) DeepSafeDrive: Machine Learning to Assess Driver Awareness

What is the leading cause of preventable death among teenagers?

Car-accidents.

In the age of texting-and-doing-everything, monitoring awareness during activities is important for safety. In this article, researchers describe a system that can use images of a driver interpreted by an algorithm to assess for spatial awareness and safety. Key points identified are the position of drivers’ hands and if they are wearing a seatbelt.

Read more here: DeepSafeDrive: A Grammar-aware Driver Parsing Approach to Driver Behavioral Situational Awareness (DB-SAW)