跳至主要内容
临床试验/CTRI/2026/01/102472
CTRI/2026/01/102472尚未招募不适用

Comprehensive hematology companion blood cell budy,a differential counter app with advanced image based

Dr. Swetha. P1 个研究点 分布在 1 个国家目标入组 100 人开始时间: 2026年2月3日最近更新:

试验速览

阶段
不适用
状态
尚未招募
发起方
入组人数
100
试验地点
1
主要终点
To calculate the differential count of each type of white blood cell using AI based image analysis of the uploaded images of microscopic fields in the blood smear

研究概览

简要总结

Blood Cell Buddy is a mobile application designed to make white blood cell differentiation faster, more standardized, and less prone to observer variation. It works by allowing users to upload or capture microscopic images of peripheral blood smears, after which the app analyzes the image and identifies various white blood cell types such as neutrophils, lymphocytes, monocytes, eosinophils, and basophils.

The algorithm uses image-based morphometric features like nuclear shape, cytoplasmic color, and granularity to distinguish between cell types. The app is structured so that a trained AI model can easily be integrated later for real diagnostic analysis. When powered by a well-trained convolutional neural network, the system can achieve accuracy levels comparable to human experts, typically exceeding 90 percent in test conditions using standardized datasets.

Clinically, this tool serves as a rapid screening aid rather than a definitive diagnostic device. It can help laboratories and clinicians perform preliminary differentials in resource-limited settings, educational demonstrations, or as a cross-check to manual counts. The automated nature of the analysis minimizes fatigue-related errors, increases reproducibility, and accelerates workflow.

By combining user-friendly interface design with future AI integration potential, the app aims to bridge the gap between manual microscopy and digital hematology, offering a reliable, accessible, and scalable approach to WBC differential counting.

研究设计

研究类型
Observational

入排标准

年龄范围
15.00 Year(s) 至 90.00 Year(s)(—)
性别
All

入选标准

  • Population criteria- Patients with various hematological derangements who undergo peripheral smear examination.
  • Medical history- Patients with a history of fever or clinical evidence of infection who undergo blood smear examination.
  • CBC Data: Patients with abnormal white blood cell distribution in automated analyzers.
  • Ethnicity and demographics: Consideration of diverse ethnic and demographic backgrounds to ensure the generalizability of the predictive model.

排除标准

  • Newborns: Excluding newborns to eliminate misdiagnosis due to the presence of nucleated red blood cells that may mimic white blood cells.
  • Lysed blood samples: Excluding patients whose samples are lysed because cells are not visualized adequately on the blood smear.
  • Inadequate clinical and lab data: Patients with inadequate clinical data are excluded, as there is no grounds to decide whether blood smear examination is warranted.

结局指标

主要结局

To calculate the differential count of each type of white blood cell using AI based image analysis of the uploaded images of microscopic fields in the blood smear

时间窗: 24 hours

次要结局

  • To calculate the accuracy and clinical utility of the AI model by correlating with manual differential counts.(1 week)

研究者

发起方
Dr. Swetha. P
申办方类型
Other [self]
责任方
Principal Investigator
主要研究者

Swetha P

Saveetha medical college and hospital, Saveetha institute of medical and technical sciences.

研究点 (1)

Loading locations...

相似试验