SHONIT™ : A powerful yet simple tool for Pathologists

separator

I am a Consultant Pathologist and  since the last 9 years I have been working in the field of Hematology . In this last decade, I have seen several paradigm shift in the field of diagnostic pathology including a shift from conventional microscopy to automated analysers. Automation has taken several quantum leaps, right from its robustness to complexity in its methodology for identifying and classifying blood cell population. Instrumentation and its associated solutions, defining the cellular components of blood, have shown several promises for the future. However, inspite of all the improvements , good old  microscopy still remains a gold standard ,because in Pathology its always “ seeing is believing”. As a Pathologist I will be more confident in an investigative modality which can give a literal insight in the cellular morphology and offer a solution based on visual analysis of images.

Today I would like to share my brief experience with Shonit, an automated solution powered by AI (Artificial Intelligence) for analysing Romanowsky stained (Leishman/ MGG) peripheral blood smears. After using Shonit™ for the first time, I had a mixed feeling of excitement tinged with apprehension and misgiving. However, gradually as I became familiar with the technology and the logical approach in designing the solution by an impressive and dedicated team of engineers and data science enthusiasts , I became more convinced in its ingenuity and in its usefulness . The solution functions simulating the human perception and analysis of  peripheral blood smears.

So for a Pathologist the obvious question would be how Shonit holds an edge over the conventional  analysers ,in reporting routine hematological parameters.

The usefulness and benefits of Shonit are manifold :

It spares a Pathologist/ technical staff from the monotony of doing a differential count of each and every slide with a manual counting chamber . This limits the observer to counting upto 100 white blood cells and repeating the process again, by manually readjusting the counter. Shonit by capturing 120 FOVs (fields of vision) not only ensures counting of more than 200-250 WBCs but also categorises them into subpopulations with fair amount of precision. Besides, it detects and enumerates those sparsely occurring WBCs which we often miss out or ignore in a manual differential counts (like basophils). A remarkable feature which I found to be very helpful is to identify and enumerate nucleated RBCs  and IGs without interfering with the counts of  routine conventional differential leucocyte parameters. Shonit is presently successfully classifying RBCs into normo, micro and macrocytes and impressively subclassifying poikilocytes.

Very soon the full array of cell identification (RBC, WBC, Platelets) will be at disposal and I am positive that Shonit will completely revolutionise the sphere of reporting in hematology.