Abstract

  In this study, we evaluate the performance of SigTuple AI100, powered by SHONITTM, a cloud based artificial intelligence (AI) system for the analysis of peripheral blood smears against the haematology automated analyser and results of manual microscopy in a large hospital setup. A blinded, randomized comparative study between results of AI100 and 7-part haematology analyser values for WBC differential counts, WBC morphological classification, RBC morphology and platelet morphology was conducted. The results of RBC and platelet morphology classification given by AI100 were compared with the results of the manual microscopy. The mean-absolute-difference between 5 Part differentials reported by AI100 and the Sysmex 7-Part haematology analyser for Neutrophil, Lymphocyte, Monocyte, Eosinophil and Basophil were 4.38%, 4.74%, 5.82%, 0.97%, and 0.53% respectively. The r2 coefficient mean of results of haematology analyser vs Shonit (AI100) for neutrophil, lymphocyte and eosinophil were 0.97, 0.97 and 0.92 respectively. AI100 can increase the throughput and decrease TAT; thereby increasing the productivity and efficiency of the pathologist.

Keywords: SigTuple AI100; RBC; WBC; Platelet

References

1. Cornet E., et al. “Performance evaluation and relevance of the CellaVisionTM DM96 system in routine analysis and in patients with malignant hematological diseases”. International Journal of Laboratory Haematology 30 (2008): 536-542.
2. Cellavision Sweden Cellavision. DM9600.
3. Medica Corporation. Easycell. USA: Medica Corporation. Medica Easycell.
4. West Medica. Vision Hema Ultimate. Austria.
5. SigTuple. ShonitTM: The complete peripheral bloodsmear analysis solution. India: SigTuple (2006).
6. Sysmex. USA (2017) Sysmex. XN-3000TM Hematology System (2017).
7. Angulo J and Flandrin G. “Automated detection of working area of peripheral blood smears using mathematical morphology”. Analytical cellular pathology 25 (2003): 37-49.
8. Ritter N and Cooper J. “Segmentation and border identification of cells in images of peripheral blood smear slides”. Australian Computer Society (2007): 161-69.
9. Bengio Y. “Learning deep architectures for AI”. Foundations and Trends in Machine Learning 2 (2009): 1-127.

Citation

Citation: Renu Ethirajan and Roja Ramani. “Study on the Performance of SigTuple AI100 in the Analysis of RBC, WBC, Platelet Morphology and WBC Differential Count in a Large Hospital Setup". Acta Scientific Medical Sciences 4.1 (2020): 102-107.

Copyright

Copyright: © 2020 Renu Ethirajan and Roja Ramani. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.