Differential diagnosis of pleural mesothelioma using Logic Learning Machine

Our plain-English summary of Differential diagnosis of pleural mesothelioma using Logic Learning Machine, by S. Parodi, R. Filiberti, P. Marroni, et al (BMC Bioinformatics, 2015).

The study

• Authors: S. Parodi, R. Filiberti, P. Marroni, R. Libener, G.P. Ivaldi, M. Mussap, E. Ferrari, C. Manneschi, E. Montani, M. Muselli
• Journal: BMC Bioinformatics
• Published: 2015;16 Suppl 9(Suppl 9):S3
• Cited by: 19 articles, as counted by NIH iCite
• Read it: PubMed · Publisher · Free full text

Abstract

Background: Tumour markers are standard tools for the differential diagnosis of cancer. However, the occurrence of nonspecific symptoms and different malignancies involving the same cancer site may lead to a high proportion of misclassifications. Classification accuracy can be improved by combining information from different markers using standard data mining techniques, like Decision Tree (DT), Artificial Neural Network (ANN), and k-Nearest Neighbour (KNN) classifier. Unfortunately, each method suffers from some unavoidable limitations. DT, in general, tends to show a low classification performance, whereas ANN and KNN produce a "black-box" classification that does not provide biological information useful for clinical purposes.

Methods: Logic Learning Machine (LLM) is an innovative method of supervised data analysis capable of building classifiers described by a set of intelligible rules including simple conditions in their antecedent part. It is essentially an efficient implementation of the Switching Neural Network model and reaches excellent classification accuracy while keeping low the computational demand. LLM was applied to data from a consecutive cohort of 169 patients admitted for diagnosis to two pulmonary departments in Northern Italy from 2009 to 2011. Patients included 52 malignant pleural mesotheliomas (MPM), 62 pleural metastases (MTX) from other tumours and 55 benign diseases (BD) associated with pleurisies. Concentration of three tumour markers (CEA, CYFRA 21-1 and SMRP) was measured in the pleural fluid of each patient and a cytological examination was also carried out. The performance of LLM and that of three competing methods (DT, KNN and ANN) was assessed by leave-one-out cross-validation.

Results: LLM outperformed all other considered methods. Global accuracy was 77.5% for LLM, 72.8% for DT, 54.4% for KNN, and 63.9% for ANN, respectively. In more details, LLM correctly classified 79% of MPM, 66% of MTX and 89% of BD. The corresponding figures for DT were: MPM = 83%, MTX = 55% and BD = 84%; for KNN: MPM = 58%, MTX = 45%, BD = 62%; for ANN: MPM = 71%, MTX = 47%, BD = 76%. Finally, LLM provided classification rules in a very good agreement with a priori knowledge about the biological role of the considered tumour markers.

Conclusions: LLM is a new flexible tool potentially useful for the differential diagnosis of pleural mesothelioma.

In plain English

This study tested whether a computer method called Logic Learning Machine (LLM) could help tell malignant pleural mesothelioma apart from other causes of fluid around the lungs. The data came from 169 patients admitted for diagnosis to two pulmonary departments in Northern Italy from 2009 to 2011: 52 with malignant pleural mesothelioma, 62 with pleural metastases from other tumours, and 55 with benign diseases. Each patient's pleural fluid was tested for three tumour markers (CEA, CYFRA 21-1 and SMRP) and examined for cancer cells. The authors compared LLM with three other methods. Overall accuracy was 77.5% for LLM, 72.8% for Decision Tree, 54.4% for k-Nearest Neighbour and 63.9% for Artificial Neural Network. LLM correctly classified 79% of mesotheliomas. Its rules for mesothelioma involved high SMRP and low CEA. The authors concluded that LLM is a flexible and powerful tool for this task and that larger studies are needed. They noted the small sample size and the low accuracy for metastases. The study did not examine asbestos exposure.

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 classification study comparing a rule-based machine-learning method (Logic Learning Machine) with three other methods (Decision Tree, k-Nearest Neighbour, Artificial Neural Network) on data from a consecutive patient cohort, assessed by leave-one-out cross-validation
• Who was studied: 169 patients admitted for diagnosis to two pulmonary departments in Northern Italy from 2009 to 2011 (52 malignant pleural mesothelioma, 62 pleural metastases from other tumours, 55 benign diseases with pleurisy). Of 177 eligible patients, 8 were excluded for missing tumour-marker data.
• Main results:
  • Overall classification accuracy across all 169 patients was 77.5% for LLM, 72.8% for Decision Tree, 54.4% for k-Nearest Neighbour and 63.9% for Artificial Neural Network
  • LLM correctly classified 79% of mesothelioma patients, 66% of metastasis patients and 89% of benign disease patients
  • Decision Tree correctly classified 83% of mesothelioma patients (slightly more than LLM), 55% of metastasis patients and 84% of benign disease patients
  • For LLM, accuracy for the two cancers taken together (mesothelioma plus metastases, the 'pooled sensitivity') was 85.1%
  • Run on the whole dataset, LLM produced 29 rules, of which 15 had very low covering (under 20%)
  • Four main LLM rules pointed to mesothelioma, each covering about 54% to 87% of mesothelioma patients. All of them involved high SMRP and low CEA, at differing cut-off values.
• The authors' conclusion: The authors concluded that LLM is a flexible and powerful tool for the differential classification of malignant mesothelioma patients. They said further studies on larger cohorts are needed to obtain stable and reproducible rules.
• Limitations the authors state:
  • The sample size was rather small
  • The comparison between the methods may have been influenced by the size of the patient groups
  • Accuracy for classifying metastases from other tumours was low
  • The LLM rules do not cover every possible combination of tumour-marker and cytology results, which could make some additional patients hard to classify

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, Pleural Mesothelioma, Soluble Mesothelin-Related Peptides (SMRP).

Related studies

• Predicting cancer using supervised machine learning: Mesothelioma — A. Choudhury (2021)
• Prediction of Radiation Pneumonitis With Machine Learning in Stage III Lung Cancer: A Pilot Study — M. Yakar, D. Etiz, et al. (2021)
• Downregulated microRNAs in the differential diagnosis of malignant pleural mesothelioma — G.V. Gee, D.C. Koestler, et al. (2010)
• Artificial intelligence-based diagnosis of asbestosis: analysis of a database with applicants for asbestosis state aid — K.B.W. Groot Lipman, C.J. de Gooijer, et al. (2023)
• The use of a next-generation sequencing-derived machine-learning risk-prediction model (OncoCast-MPM) for malignant pleural mesothelioma: a retrospective study — M.G. Zauderer, A. Martin, et al. (2021)
• Value of p63 and cytokeratin 5/6 as immunohistochemical markers for the differential diagnosis of poorly differentiated and undifferentiated carcinomas — O. Kaufmann, E. Fietze, et al. (2001)

Browse return to medical research.