| 1. |
- Bokhorst, John-Melle, et al.
(författare)
-
Deep learning for multi-class semantic segmentation enables colorectal cancer detection and classification in digital pathology images
- 2023
-
Ingår i: Scientific Reports. - : NATURE PORTFOLIO. - 2045-2322. ; 13:1
-
Tidskriftsartikel (refereegranskat)abstract
- In colorectal cancer (CRC), artificial intelligence (AI) can alleviate the laborious task of characterization and reporting on resected biopsies, including polyps, the numbers of which are increasing as a result of CRC population screening programs ongoing in many countries all around the globe. Here, we present an approach to address two major challenges in the automated assessment of CRC histopathology whole-slide images. We present an AI-based method to segment multiple (n=14 ) tissue compartments in the H &E-stained whole-slide image, which provides a different, more perceptible picture of tissue morphology and composition. We test and compare a panel of state-of-the-art loss functions available for segmentation models, and provide indications about their use in histopathology image segmentation, based on the analysis of (a) a multi-centric cohort of CRC cases from five medical centers in the Netherlands and Germany, and (b) two publicly available datasets on segmentation in CRC. We used the best performing AI model as the basis for a computer-aided diagnosis system that classifies colon biopsies into four main categories that are relevant pathologically. We report the performance of this system on an independent cohort of more than 1000 patients. The results show that with a good segmentation network as a base, a tool can be developed which can support pathologists in the risk stratification of colorectal cancer patients, among other possible uses. We have made the segmentation model available for research use on .
|
|
| 2. |
- Caputo, Alessandro, et al.
(författare)
-
Digital Examination of LYmph node CYtopathology Using the Sydney system (DELYCYUS). An international, multi-institutional study
- 2023
-
Ingår i: Cancer Cytopathology. - 1934-662X. ; 131:11, s. 679-692
-
Tidskriftsartikel (refereegranskat)abstract
- Background: After a series of standardized reporting systems in cytopathology, the Sydney system was recently introduced to address the need for reproducibility and standardization in lymph node cytopathology. Since then, the risk of malignancy for the categories of the Sydney system has been explored by several studies, but no studies have yet examined the interobserver reproducibility of the Sydney system. Methods: The authors assessed interobserver reproducibility of the Sydney system on 85 lymph node fine-needle aspiration cytology cases reviewed by 15 cytopathologists from 12 institutions in eight different countries, resulting in 1275 diagnoses. In total, 186 slides stained with Diff-Quik, Papanicolaou, and immunocytochemistry were scanned. A subset of the cases included clinical data and results from ultrasound examinations, flow cytometry immunophenotyping, and fluorescence in situ hybridization analysis. The study participants assessed the cases digitally using whole-slide images. Results: Overall, the authors observed an almost perfect agreement of cytopathologists with the ground truth (median weighted Cohen κ = 0.887; interquartile range, κ = 0.210) and moderate overall interobserver concordance (Fleiss κ = 0.476). There was substantial agreement for the inadequate and malignant categories (κ = 0.794 and κ = 0.729, respectively), moderate agreement for the benign category (κ = 0.490), and very slight agreement for the suspicious (κ = 0.104) and atypical (κ = 0.075) categories. Conclusions: The Sydney system for reporting lymph node cytopathology shows adequate interobserver concordance. Digital microscopy is an adequate means to assess lymph node cytopathology specimens.
|
|
| 3. |
- Marini, Niccolo, et al.
(författare)
-
Unleashing the potential of digital pathology data by training computer-aided diagnosis models without human annotations
- 2022
-
Ingår i: npj Digital Medicine. - : Nature Portfolio. - 2398-6352. ; 5:1
-
Tidskriftsartikel (refereegranskat)abstract
- The digitalization of clinical workflows and the increasing performance of deep learning algorithms are paving the way towards new methods for tackling cancer diagnosis. However, the availability of medical specialists to annotate digitized images and free-text diagnostic reports does not scale with the need for large datasets required to train robust computer-aided diagnosis methods that can target the high variability of clinical cases and data produced. This work proposes and evaluates an approach to eliminate the need for manual annotations to train computer-aided diagnosis tools in digital pathology. The approach includes two components, to automatically extract semantically meaningful concepts from diagnostic reports and use them as weak labels to train convolutional neural networks (CNNs) for histopathology diagnosis. The approach is trained (through 10-fold cross-validation) on 3769 clinical images and reports, provided by two hospitals and tested on over 11000 images from private and publicly available datasets. The CNN, trained with automatically generated labels, is compared with the same architecture trained with manual labels. Results show that combining text analysis and end-to-end deep neural networks allows building computer-aided diagnosis tools that reach solid performance (micro-accuracy = 0.908 at image-level) based only on existing clinical data without the need for manual annotations.
|
|
