Improving lung cancer pathological hyperspectral diagnosis through cell-level annotation refinement
Open Access
- 8 March 2025
- journal article
- research article
- Published by Springer Nature in Scientific Reports
- Vol. 15 (1) , 1-14
Abstract
Lung cancer remains a major global health challenge, and accurate pathological examination is crucial for early detection. This study aims to enhance hyperspectral pathological image analysis by refining annotations at the cell level and creating a high-quality hyperspectral dataset of lung tumors. We address the challenge of coarse manual annotations in hyperspectral lung cancer datasets, which limit the effectiveness of deep learning models requiring precise labels for training. We propose a semi-automated annotation refinement method that leverages hyperspectral data to enhance pathological diagnosis. Specifically, we employ K-means unsupervised clustering combined with human-guided selection to refine coarse annotations into cell-level masks based on spectral features. Our method is validated using a hyperspectral lung squamous cell carcinoma dataset containing 65 image samples. Experimental results demonstrate that our approach improves pixel-level segmentation accuracy from 77.33% to 92.52% with a lower level of prediction noise. The time required to accurately label each pathological slide is significantly reduced. While pixel-level labeling methods for an entire slide can take over 30 mins, our semi-automated method requires only about 5 mins. To enhance visualization for pathologists, we apply a conservative post-processing strategy for instance segmentation. These results highlight the effectiveness of our method in addressing annotation challenges and improving the accuracy of hyperspectral pathological analysis.Keywords
Funding Information
- Equipment Research Program of the Chinese Academy of Sciences (YJKYYQ20180039)
- Informatization Plan of the Chinese Academy of Sciences (CASWX2021-PY-0110)
- Natural Science Foundation of Beijing Municipality (JQ22029)
- Major Project of Guangzhou National Laboratory (GZNL2023A03009)
- Fonds Wetenschappelijk Onderzoek (1252722N, G0A3O24N, VS03924N)
This publication has 51 references indexed in Scilit:
- Empirical exploration of predictive maintenance in concrete manufacturing: Harnessing machine learning for enhanced equipment reliability in construction project managementComputers & Industrial Engineering, 2024
- Exploring the efficacy of machine learning models for predicting soil radon exhalation ratesStochastic Environmental Research and Risk Assessment, 2023
- Hyperspectral Imaging for Clinical ApplicationsBioChip Journal, 2022
- 3D‐PulCNN: Pulmonary cancer classification from hyperspectral images using convolution combination unit based CNNJournal of Biophotonics, 2021
- Cancer statistics for the year 2020: An overviewInternational Journal of Cancer, 2021
- Various dimension reduction techniques for high dimensional data analysis: a reviewArtificial Intelligence Review, 2021
- A local pre-processing method for near-infrared spectra, combined with spectral segmentation and standard normal variate transformationAnalytica Chimica Acta, 2016
- Practical issues of hyperspectral imaging analysis of solid dosage formsAnalytical and Bioanalytical Chemistry, 2010
- Use of hyperspectral imaging to distinguish normal, precancerous, and cancerous cellsCancer, 2008
- Hyperspectral imaging: A novel approach for microscopic analysisCytometry, 2001