Attention-guided multi-scale deep object detection framework for lymphocyte analysis in IHC histological images

Abstract

Tumor-Infiltrating Lymphocytes are specialized lymphocytes that can detect and kill cancerous cells. Their detection poses many challenges due to significant morphological variations, overlapping occurrence, artifact regions, and the high-class resemblance between clustered areas and artifacts. In this regard, a Lymphocyte Analysis Framework (DC-Lym-AF) based on Deep Convolutional Neural Network (CNN) is proposed to analyze lymphocytes in immunohistochemistry images. The proposed framework comprises: i) pre-processing, ii) screening phase, iii) localization phase, and iv) post-processing. In the screening phase, a custom CNN architecture (Lymph-Dil-Net) is developed to screen lymphocytic regions by performing a patch-level classification. This proposed architecture uses dilated convolutions and shortcut connections to capture multi-level variations and ensure reference-based learning. In contrast, the localization phase utilizes an Attention-guided Multi-Scale Lymphocyte Detector (Attn-MS-LD) to detect lymphocytes. The proposed detector extracts refined and multi-scale features by exploiting dilated convolutions, attention mechanism, and feature pyramid network using its custom attention-aware backbone. The proposed DC-Lym-AF shows exemplary performance on the NuClick dataset compared to existing detection models with an F-score and precision of 0.84 and 0.83, respectively. We verified the generalizability of our proposed framework by participating in a publically open LYON’19 challenge. Results in terms of detection rate (0.76) and F-score (0.73) suggest that the proposed DC-Lym-AF can effectively detect lymphocytes in IHC-stained images collected from different laboratories. In addition, its promising generalization on several datasets implies that it can be turned into a medical diagnostic tool to investigate various histopathological problems. Mini- Abstract A novel Lymphocyte Analysis Framework (DC-Lym-AF) based on Deep Convolutional Neural Network is proposed to analyze lymphocytes in immunohistochemistry images. The proposed framework comprises: i) pre-processing, ii) screening phase, iii) localization phase, and iv) post-processing. The proposed framework outperformed in detecting lymphocytes when compared with state-of-the-art detection models.

Lymphocyte detection for cancer analysis using a novel fusion block based channel boosted CNN

“…In this regard, pre-trained ImageNet weights 57 were used for the weight optimization of the comparative models. In addition, we also compared the proposed approach with a recent work, “DC-Lym-AF” 58 . Results depicted in Table 8 , show that the proposed “BCF-Lym-Detector” has detected lymphocytes with a high detection rate as compared to other existing detectors.…”

LYSTO: The Lymphocyte Assessment Hackathon and Benchmark Dataset

“…LYSTO has already supported a series studies. These works mainly focus on lymphocyte IHC scoring and use detection models such as Faster R-CNN and Mask R-CNN [44]- [46]. Inspired by LYSTO, [47] explored an interactive annotation framework.…”

Novel Deep CNNs Explore Regions, Boundaries, and Residual Learning for COVID-19 Infection Analysis in Lung CT

“…However, CT scan analysis is often slow, laborious, and susceptible to human error. Consequently, DL-based diagnostic tools have been developed to expedite and improve image analysis, aiding healthcare professionals [17]–[19]. DL techniques have demonstrated optimal performance in image analysis with deep CNNs being particularly prominent [20].…”