Computer-assisted differential diagnosis of malignant mesothelioma based on syntactic structure analysis

Our plain-English summary of Computer-assisted differential diagnosis of malignant mesothelioma based on syntactic structure analysis, by B. Weyn, G. van de Wouwer, S. Kumar-Singh, et al (Cytometry, 1999).

The study

• Authors: B. Weyn, G. van de Wouwer, S. Kumar-Singh, A. van Daele, P. Scheunders, E. van Marck, W. Jacob
• Journal: Cytometry
• Published: 1999;35(1):23-9
• Cited by: 20 articles, as counted by NIH iCite
• Read it: PubMed · Publisher

Abstract

Background: Malignant mesothelioma, a mesoderm-derived tumor, is related to asbestos exposure and remains a diagnostic challenge because none of the genetic or immunohistochemical markers have yet been proven to be specific. To assist in the identification of mesothelioma and to differentiate it from other common lesions at the same location, we have tested the performance of syntactic structure analysis (SSA) in an automated classification procedure.

Materials And Methods: Light-microscopic images of tissue sections of malignant mesothelioma, hyperplastic mesothelium, and adenocarcinoma were analyzed using parameters selected from the Voronoi diagram, Gabriel's graph, and the minimum spanning tree which were classified with a K-nearest-neighbor algorithm.

Results: Results showed that mesotheliomas were diagnosed correctly in 74% of the cases; 76% of the adenocarcinomas were correctly graded, and 88% of the mesotheliomas were correctly typed. The performance of the parameters was dependent on the obtained classification (i.e., tumor-tumor versus tumor-benign).

Conclusions: Our results suggest that SSA is valuable in the differential classification of mesothelioma and that it supplements a visually appraised diagnosis. The recognition scores may be increased by a combination of SSA with, for example, cellular or nuclear parameters, measured at higher magnifications to form a solid base for fully automated expert systems.

In plain English

This laboratory study tested whether a computer method could help tell malignant mesothelioma apart from other tissue found at the same site. The method, called syntactic structure analysis, measures how cell nuclei are arranged in a stained tissue section. The authors analyzed microscope images from archived tissue at the University Hospital of Antwerp, Belgium: 44 cases of malignant mesothelioma, 7 cases of hyperplastic mesothelium (a benign overgrowth) and 38 cases of lung carcinoma, described elsewhere in the article as adenocarcinoma of the lung. A computer classifier assigned each case to a group, using measurements combined across that case's images. Asbestos exposure was not studied. Across all three groups, 73.9% of cases were classified correctly. The figure was 81% for mesothelioma and 73% for adenocarcinoma. Hyperplastic mesothelium was often mislabeled, and 4.5% of mesotheliomas were mislabeled as hyperplastic mesothelium. The method sorted mesotheliomas into epithelial, sarcomatoid and biphasic types correctly in 88.64% of cases. It graded adenocarcinomas correctly in 75.7%. The authors concluded that the method is valuable for telling mesothelioma apart and supplements a visual diagnosis. They noted that the number of cases was limited and that there were few hyperplastic samples.

We wrote this plain-English summary from the full article with the help of AI, and a second AI model checked each sentence against the article. It is not the authors' wording; the abstract and the article are their own account.

What the study did and found

• Design: Diagnostic image-analysis study of archived tissue sections. Computer-measured tissue-architecture parameters, combined per case across that case's images, were classified with a K-nearest-neighbor classifier.
• Who was studied: Archived formalin-fixed, paraffin-embedded tissue from 89 cases at the Laboratory of Pathology, University Hospital of Antwerp, Belgium: 44 malignant mesothelioma, 7 hyperplastic mesothelium, 38 pulmonary carcinomas (adenocarcinoma of the lung)
• Exposure: No asbestos exposure was measured. The article states only as background that mesothelioma is linked to asbestos exposure. What was studied was a computer image-analysis method (syntactic structure analysis).
• Main results:
  • In distinguishing adenocarcinoma, malignant mesothelioma and hyperplastic mesothelium, 73.9% of cases overall were correctly classified
  • The recognition score was 73% for adenocarcinoma and 81% for malignant mesothelioma (81.8% in Table 1). When these two were misclassified, it was usually as each other.
  • Misclassification occurred mainly with hyperplastic mesothelium, which was mislabeled both as malignant mesothelioma and as adenocarcinoma. 4.5% of malignant mesotheliomas were mislabeled as hyperplastic mesothelium.
  • Typing of malignant mesothelioma as epithelial, sarcomatoid or biphasic was correct in 88.64% of mesothelioma cases. Misclassification was almost entirely biphasic cases labeled as epithelial.
  • Epithelial and sarcomatoid mesotheliomas were classified correctly in 93% and 100%, respectively
  • Grading of lung adenocarcinoma was correct in 75.7% of adenocarcinoma cases
• The authors' conclusion: The authors concluded that syntactic structure analysis is valuable in the differential classification of mesothelioma and supplements a visually appraised diagnosis. They added that combining it with cellular or nuclear measurements may increase recognition scores.
• Limitations the authors state:
  • The number of cases (89) was limited compared with the number of parameters
  • The number of hyperplastic mesothelium samples was small, which partly explains their misclassification
  • In very complex cell clusters, some nuclei were missed by the image analysis

Topics

The National Library of Medicine indexes this study under subjects our site covers: Lung cancer, Mesothelioma.

Medical terms in this study, explained in our Lexicon: Differential Diagnosis, Malignant Mesothelioma, Mesoderm, Asbestos Exposure, Mesothelioma, Mesothelium, Adenocarcinoma.

Related studies

• Computer-assisted diagnosis for early stage pleural mesothelioma: towards automated detection and quantitative assessment of pleural thickening from thoracic CT images — K. Chaisaowong, T. Aach, et al. (2007)
• Value of morphometry, texture analysis, densitometry, and histometry in the differential diagnosis and prognosis of malignant mesothelioma — B. Weyn, G. Van De Wouwer, et al. (1999)
• Computer-assisted classification system for chest X-ray and computed tomography findings in occupational lung disease — T. Kraus, H.J. Raithel, et al. (1997)
• Value of Glut-1 and Koc markers in the differential diagnosis of reactive mesothelial hyperplasia, malignant mesothelioma and pulmonary adenocarcinoma — Ö. Üçer, A.F. Dağli, et al. (2013)
• Use of antimesothelial cell antibody and computer assisted quantitative analysis for distinguishing between reactive and neoplastic serosal tissues — A. Donna, P.G. Betta, et al. (1987)
• Use of intermediate filament antibodies in the differential diagnosis of gynecological neoplasia — F.C. Ramaekers, J.J. Puts, et al. (1985)

Browse return to medical research.