A breakthrough in the diagnostic challenge of monocytic neoplasms with the use of Shonit™ powered by artificial intelligence.
Detection of IMMs (Immature monocytes) – Artificial intelligence versus Hematology analyzers
by Dr. Renu Ethirajan
Diagnosis of monocytic neoplasms have always been a challenge in haemato-oncology. Morphological assessment is subject to significant intra-observer variability. Accurate diagnosis of Acute Myeloid Leukemia (AML) with monocytic differentiation has always been daunting for most haemato-pathologists. While cytochemical tests can help in certain cases, only the flow cytometry technique using anti-CD14, antibodies have been proven to be useful for morphologically difficult cases. In addition, the use of MO2 and MY4 anti-CD14 antibodies have demonstrated their usefulness in evaluating suspected cases of AMLs with monocytic differentiation.
Although flow cytometric analysis may be essentially useful in distinguishing the maturation pattern in monocytes, not many labs and hospitals can afford this expensive methodology.
Accurate percentages of monoblasts, promonocytes and monocytes need to be determined to distinguish between AML with monocytic differentiation and other Chronic Myeloproliferative Disorders (CMPDs) such as CMML.
Recently, haematology analyzers have come up with a new research parameter such as “IMM” which highlights the immature monocyte population in the sample. This technique is based on the principle of light scatter and fluorescence. For the practicing hemato-pathologist, this is still a “flag” given by the analyzer which needs to be confirmed by manual microscopy.
In a country like India where most labs in the periphery as well as urban cities still have the 3 – part haematology analyzers, such research parameters are far from reach. Even if they are available, it would still mandate a manual confirmation by the conventional microscopy followed by flow cytometry for further characterization.
Leukaemias fall under critical reporting and require an “ASAP” diagnosis especially when the clinical presentation involves bleeding or disseminated intravascular coagulation – like symptoms.
In these situations, artificial intelligence powered solutions such as Shonit™ prove as an aid to the practicing haemato-oncopathologist. Its ability to provide standardized and accurate classification of monocytes, promonocytes and monoblasts, developed on a background of deep learning will enable the pathologist to compute the differential counts with ease and form a reliable diagnosis. The supportive visual evidences provided by the software enhances the confidence level of the report.
Visual evidences of blasts with monocytic differentiation provided by Shonit™
Screening of leukemic cells has never been much easier. Since the solution enables telepathology, an expert opinion after screening can be obtained within a short time enabling quicker decisions in treatment protocols.
Shonit™ qualifies as a powerful screening and diagnostic tool for leukemia with additional benefits of classification resulting from an in-depth analysis of the atypical cells and blasts. A thorough quantitation of blasts along with its precision in sub-typing makes it an excellent ‘one-stop’ AI driven solution to meet the diagnostic challenges of haematological malignancies.
Clearly AI is the need of the hour, especially in oncology where the timing of diagnosis is crucial. Through this solution offered by Shonit™, a pathologist can be most certain of achieving the accuracy, speed and confidence in the diagnosis of haematological malignancies.