Computer-aided volumetric assessment of malignant pleural mesothelioma on CT using a random walk-based method

Our plain-English summary of Computer-aided volumetric assessment of malignant pleural mesothelioma on CT using a random walk-based method, by M. Chen, E. Helm, N. Joshi, et al (International Journal of Computer Assisted Radiology and Surgery, 2017).

The study

• Authors: M. Chen, E. Helm, N. Joshi, F. Gleeson, M. Brady
• Journal: International Journal of Computer Assisted Radiology and Surgery
• Published: 2017;12(4):529-538
• Cited by: 9 articles, as counted by NIH iCite
• Read it: PubMed · Publisher · Free full text

Abstract

Objective: The aim of this study is to assess the performance of a computer-aided semi-automated algorithm we have adapted for the purpose of segmenting malignant pleural mesothelioma (MPM) on CT.

Methods: Forty-five CT scans were collected from 15 patients (M:F [Formula: see text] 10:5, mean age 62.8 years) in a multi-centre clinical drug trial. A computer-aided random walk-based algorithm was applied to segment the tumour; the results were then compared to radiologist-drawn contours and correlated with measurements made using the MPM-adapted Response Evaluation Criteria in Solid Tumour (modified RECIST).

Results: A mean accuracy (Sørensen-Dice index) of 0.825 (95% CI [0.758, 0.892]) was achieved. Compared to a median measurement time of 68.1 min (range [40.2, 102.4]) for manual delineation, the median running time of our algorithm was 23.1 min (range [10.9, 37.0]). A linear correlation (Pearson's correlation coefficient: 0.6392, [Formula: see text]) was established between the changes in modified RECIST and computed tumour volume.

Conclusion: Volumetric tumour segmentation offers a potential solution to the challenges in quantifying MPM. Computer-assisted methods such as the one presented in this study facilitate this in an accurate and time-efficient manner and provide additional morphological information about the tumour's evolution over time.

In plain English

This study tested a computer program meant to help measure malignant pleural mesothelioma tumours on CT scans. It did not examine asbestos exposure. The authors used CT scans from 15 patients (10 men, 5 women, mean age 62.8) treated at the UK centres of a Phase II drug trial of vinflunine. The article gives the number of scans as 45 in some places and 48 in another. A radiologist placed starting points ("seeds") on some scan slices, and the program then outlined the tumour in three dimensions. Its outlines were compared with outlines a radiologist drew by hand. Overlap with the hand-drawn outlines, measured by the Sørensen–Dice index, averaged 0.825 (95% CI 0.758 to 0.892). The program's median running time was 23.1 minutes, compared with 68.1 minutes for drawing by hand. Changes in computed tumour volume correlated with changes in modified RECIST measurements (Pearson's coefficient 0.6392). The authors concluded that the results support using the method for diagnostic and treatment-monitoring purposes. They noted that their reference outlines might not be ideal and that larger datasets are needed.

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: Technical evaluation of a semi-automated computer method for outlining (segmenting) mesothelioma tumours on CT, using scans collected in a Phase II drug trial and compared with a radiologist's hand-drawn outlines
• Who was studied: 15 patients with histologically and cytologically confirmed mesothelioma from the UK centres of a multi-centre Phase II vinflunine trial run across nine centres in the UK, France and Germany. There were 10 men and 5 women, with a mean age of 62.8 years (range 47.9 to 79.7). The methods section gives 48 baseline and follow-up CT scans; the abstract and Table 4 give 45.
• Exposure: Not an exposure study; asbestos exposure was not assessed. What was evaluated was a semi-automatic random walk-based segmentation algorithm, extended to 3-D, applied to non-contrast CT scans.
• Main results:
  • Across the patient data, the program's outlines agreed with the radiologist's hand-drawn outlines with a mean Sørensen–Dice index of 0.825 (95% CI 0.758 to 0.892).
  • Drawing the tumour by hand took a median of 68.1 minutes per scan (range 40.2 to 102.4). The computer-aided procedure took a median of 23.1 minutes (range 10.9 to 37.0).
  • The user's hands-on interaction with the program took a median of 2.8 minutes per scan (range 1.8 to 4.2).
  • Changes in computed tumour volume were correlated with changes in modified RECIST measurements: Pearson's correlation coefficient 0.6392, p value 0.0001, R-squared 0.4086.
  • Changing the placement of the starting seeds did not significantly change overall accuracy, but accuracy generally fell with distance from the slices where seeds were placed.
  • Extra seeds had to be added at a review stage in one out of every three scans.
• The authors' conclusion: The authors concluded that their results support using the algorithm to segment malignant pleural mesothelioma for both diagnostic and treatment-monitoring purposes.
• Limitations the authors state:
  • The reference outlines were drawn by one radiologist. The authors note that such outlines vary between observers and over time and might not be ideal as a gold standard.
  • The dataset was too small to validate volume-based response thresholds: only 1 of the 15 patients would be classified as having progressive disease and the rest as stable disease.
  • The authors state that larger patient datasets and more detailed follow-up clinical data are needed.

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: Pleural Mesothelioma, CT Scan, Modified RECIST (mRECIST).

Related studies

• Assessment of therapy responses and prediction of survival in malignant pleural mesothelioma through computer-aided volumetric measurement on computed tomography scans — F. Liu, B. Zhao, et al. (2010)
• 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)
• 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)
• Three-dimensional stereoscopic volume rendering of malignant pleural mesothelioma — N.M. Mollberg, N.M. Parsad, et al. (2012)
• Sample preparation method for visualization of nanoparticulate captured on mixed cellulose ester filter media by enhanced darkfield microscopy and hyperspectral imaging — N.M. Neu-Baker, A.C. Eastlake, et al. (2019)
• PET-guided dose escalation tomotherapy in malignant pleural mesothelioma — A. Fodor, C. Fiorino, et al. (2011)

Browse return to medical research.