Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Large-scale Multi Domain Leukemia Dataset for the White Blood Cells Detection with Morphological Attributes for Explainability

Abdul Rehman, Talha Meraj, Aiman Mahmood Minhas, Ayisha Imran, Mohsen Ali, Waqas Sultani

TL;DR

The paper tackles the difficulty of early leukemia prognosis by leveraging morphological information of white blood cells (WBCs) and addressing data scarcity through a large, multi-domain dataset. It introduces LeukemiaAttri, a multi-domain WBC dataset captured across high-cost and low-cost microscopes at 10x, 40x, and 100x magnifications, with localization and annotations for 14 WBC types and 7 morphological attributes. To enable explainable prognosis, it proposes AttriDet, a multi-headed detector based on YOLOv5 that jointly predicts WBC types and morphological attributes, aided by a morphology bank and text-based cell descriptions. Extensive experiments across 12 domain subsets evaluate object detection, unsupervised domain adaptation, and attribute prediction, showing strong performance for YOLOv5 variants, limited but encouraging gains from UDA, and clear advantages of incorporating attribute prediction, especially in challenging attributes like nucleus details and cytoplasmic vacuoles.

Abstract

Earlier diagnosis of Leukemia can save thousands of lives annually. The prognosis of leukemia is challenging without the morphological information of White Blood Cells (WBC) and relies on the accessibility of expensive microscopes and the availability of hematologists to analyze Peripheral Blood Samples (PBS). Deep Learning based methods can be employed to assist hematologists. However, these algorithms require a large amount of labeled data, which is not readily available. To overcome this limitation, we have acquired a realistic, generalized, and large dataset. To collect this comprehensive dataset for real-world applications, two microscopes from two different cost spectrums (high-cost HCM and low-cost LCM) are used for dataset capturing at three magnifications (100x, 40x, 10x) through different sensors (high-end camera for HCM, middle-level camera for LCM and mobile-phone camera for both). The high-sensor camera is 47 times more expensive than the middle-level camera and HCM is 17 times more expensive than LCM. In this collection, using HCM at high resolution (100x), experienced hematologists annotated 10.3k WBC types (14) and artifacts, having 55k morphological labels (Cell Size, Nuclear Chromatin, Nuclear Shape, etc.) from 2.4k images of several PBS leukemia patients. Later on, these annotations are transferred to other 2 magnifications of HCM, and 3 magnifications of LCM, and on each camera captured images. Along with the LeukemiaAttri dataset, we provide baselines over multiple object detectors and Unsupervised Domain Adaptation (UDA) strategies, along with morphological information-based attribute prediction. The dataset will be publicly available after publication to facilitate the research in this direction.

A Large-scale Multi Domain Leukemia Dataset for the White Blood Cells Detection with Morphological Attributes for Explainability

TL;DR

The paper tackles the difficulty of early leukemia prognosis by leveraging morphological information of white blood cells (WBCs) and addressing data scarcity through a large, multi-domain dataset. It introduces LeukemiaAttri, a multi-domain WBC dataset captured across high-cost and low-cost microscopes at 10x, 40x, and 100x magnifications, with localization and annotations for 14 WBC types and 7 morphological attributes. To enable explainable prognosis, it proposes AttriDet, a multi-headed detector based on YOLOv5 that jointly predicts WBC types and morphological attributes, aided by a morphology bank and text-based cell descriptions. Extensive experiments across 12 domain subsets evaluate object detection, unsupervised domain adaptation, and attribute prediction, showing strong performance for YOLOv5 variants, limited but encouraging gains from UDA, and clear advantages of incorporating attribute prediction, especially in challenging attributes like nucleus details and cytoplasmic vacuoles.

Abstract

Earlier diagnosis of Leukemia can save thousands of lives annually. The prognosis of leukemia is challenging without the morphological information of White Blood Cells (WBC) and relies on the accessibility of expensive microscopes and the availability of hematologists to analyze Peripheral Blood Samples (PBS). Deep Learning based methods can be employed to assist hematologists. However, these algorithms require a large amount of labeled data, which is not readily available. To overcome this limitation, we have acquired a realistic, generalized, and large dataset. To collect this comprehensive dataset for real-world applications, two microscopes from two different cost spectrums (high-cost HCM and low-cost LCM) are used for dataset capturing at three magnifications (100x, 40x, 10x) through different sensors (high-end camera for HCM, middle-level camera for LCM and mobile-phone camera for both). The high-sensor camera is 47 times more expensive than the middle-level camera and HCM is 17 times more expensive than LCM. In this collection, using HCM at high resolution (100x), experienced hematologists annotated 10.3k WBC types (14) and artifacts, having 55k morphological labels (Cell Size, Nuclear Chromatin, Nuclear Shape, etc.) from 2.4k images of several PBS leukemia patients. Later on, these annotations are transferred to other 2 magnifications of HCM, and 3 magnifications of LCM, and on each camera captured images. Along with the LeukemiaAttri dataset, we provide baselines over multiple object detectors and Unsupervised Domain Adaptation (UDA) strategies, along with morphological information-based attribute prediction. The dataset will be publicly available after publication to facilitate the research in this direction.
Paper Structure (9 sections, 3 figures, 4 tables)

This paper contains 9 sections, 3 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Dataset
  3. LeukemiaAttri Dataset
  4. AttriDet: A WBC detection method with attributes
  5. Experiments
  6. Object Detection
  7. Unsupervised Domain Adaptation based Object detection
  8. Object detection with attribute prediction
  9. Conclusions

Figures (3)

  • Figure 1: a) Illustrate the image-capturing procedure using general mobile cameras (C1) and high-end cameras for HCM, and middle-level cameras for LCM (C2) at multiple resolutions. Images are captured using both high and low-cost microscopes. b) shows our microscope set up with cameras (C1, C2) to capture slide images and cameras (C3, C4) to capture the stage-scales. c) Shows the different types of WBCs with morphological attributes.
  • Figure 2: (a): WBC type and (b) Morphological attributes distribution
  • Figure 3: AttriDet: WBC and their corresponding attribute predictions framework