Indo-American scientists develop algorithm for early detection of Brain Blood Barrier

HYDERABAD: Two Indo-American Scientists, Venkata Ravi Kiran Kolla, a senior software engineer and senior research scientist, along with Niharika Reddy Meenigea, researcher and a senior data analyst, developed and proposed a decision tree-based classification algorithm for the early detection of blood-brain barrier (BBB).

The blood-brain barrier is dangerously emerging as one of the fatal neurodegenerative disorders, depicting itself as a progressive degeneration of cognitive abilities.

Seventy-Six million people, most of whom are above 65 years of age, are affected by BBB worldwide. It has also been reported that around 10% of BBB patients are within the age group of 30 years to 60 years. It is regarded as the most common form of dementia; presently, 60 - 70% of dementia is due to BBB.

BBB is caused primarily due to the loss of brain tissues like white matter (WM), gray matter (GM), etc., and this BBB enhances the volume of cerebra spinal fluid (CSF). Symptoms of BBB range from forgetfulness in its early stage to loss of speech and cognitive ability in a later stage. Sometimes the symptoms of BBB are mistakenly ignored even by the friends and relatives of the patient. Generally, the transition of a healthy patient to BBB occurs through an intermediate stage called mild cognitive impairment (MCI).

Scientists have researched other fields of medical science and found an alternate solution that the Machine Learning technique is the easiest and best way to reduce the risks.

Therefore, early detection of BBB has surfaced as a need of the hour, as presently, no clinical diagnosis is standardized. In this regard, Scientists Niharikareddy Meenigea and Venkata Ravi Kiran Kolla have worked together to deliver a timely solution.

The scientists have developed machine learning-based algorithms which have received justified interest among researchers across different disciplines, and the field of medical image processing is no exception.

According to the research, the Performance of this decision tree-based classification process is subsequently measured using relevant metrics like accuracy, sensitivity, and specificity.