Gamma

Radionuclide medical diagnostics detects how well internal organs function. The most common techniques are SPECT Tomography and Gamma Camera scans. Recording the radiation emitted by an injected radioactive tracer from within the body, they construct a 3D image – dynamic scintigram – showing how well the organs work. We used AI to analyse these images.

The Problem

Traditionally, analysis of heart, lungs and urinary tract medical images was exclusively a trained doctor’s domain. The process places a high cognitive load on the specialist and is costly, inefficient and slow.

Our Solution

Having analysed the client’s dataset, we proposed an intelligent decision support system. At the core of our AI was a proprietary modification of swarming algorithm used for informative feature selection and an information-extreme classifier with complex class containers.

Resulting intelligent advisor was capable of:

01

Processing images of heart, lungs and kidneys, automatically segmenting them to determine areas of interest requiring specific attention

02

Detecting parenchymal kidney disease by automatically classifying the data in the areas of interest in renal scintigrams

03

Diagnosing dyskinetic abnormalities and substantial damage to lungs

04

Performing automatic analysis of the myocardial perfusion imaging to diagnose various forms of ischemic heart disease and detect heart scar tissue

The Outcome

Our intelligent decision support system helped automate a number of cognitive-repetitive tasks, reducing the cognitive demands on the medical professionals. Innovative use of proprietary AI technologies resulted in 99% average decision accuracy, supporting the doctors in their decision-making and allowing them to zero in on the more difficult cases where only a human expertise would do.