A Machine Learning-Based Approach for Classification of Focal Splenic Lesions Based on Their CT Features

Abstract

The aim of the study was to describe the CT features of focal splenic lesions (FSLs) in dogs in order to predict lesion histotype. Dogs that underwent a CT scan and had a FSL diagnosis by cytology or histopathology were retrospectively included in the study. For the statistical analysis the cases were divided into four groups, based on the results of cytopatholoy or hystopathology, namely: nodular hyperplasia (NH), other benign lesions (OBLs), sarcoma (SA), round cell tumour (RCT). Several qualitative and quantitative CT features were described for each case. The relationship occurring between each individual CT feature and the histopathological groups was explred by means of c chi-square test for the count data and by means of Kruskal-Wallis or ANOVA for the continuous data. Furthermore, the main features of each group were described using factorial discriminant analysis, and a decision tree for lesion classification was then developed. Sarcomas were characterised by large dimensions, a cystic appearance and an overall low post contrast-enhancement. NH and OBLs were characterised by small dimensions, a solid appearance and a high post-contrast enhancement. OBLs showed higher post-contrast values than NH. Lastly, RCTs did not exhibit any distinctive CT features. The proposed decision tree had a high accuracy for the classification of SA (0.89) and a moderate accuracy for the classification of OBLs and NH (0.79), whereas it was unable to classify RCTs. The results of the factorial analysis and the proposed decision tree could help the clinician in classifying FSLs based on their CT features. A definitive FSL diagnosis can only be obtained by microscopic examination of the spleen.

Pathological Characterization and Risk Factors of Splenic Nodular Lesions in Dogs (Canis lupus familiaris)

“…Sarcomas are usually large, have a cystic appearance, and have low post-contrast enhancement. On the other hand, benign lesions are small, solid, and have high post-contrast enhancement [ 38 ].…”

Histological interpretation of spitzoid tumours: an extensive machine learning‐based concordance analysis for improving decision making

“…17 Additionally, the ANOVA analysis has proved to be a valuable tool for identifying and selecting the most significant features, enhancing the performance of ML models to achieve greater accuracy in various applications related to tumour diagnosis. [18][19][20][21] As spitzoid tumours continue to pose significant challenges in dermatopathology, this study aimed to evaluate the effectiveness of ML models in distinguishing benign from malignant tumours, as well as predicting the subclassification of the atypical intermediate category, based on 22 clinicopathological features in different cohorts diagnosed by dermatopathologists from four different countries. The primary goal was to rank these features objectively according to their relevance, providing valuable insights to pathologists in identifying the most significant factors for accurate diagnosis.…”

Contrast-enhanced ultrasound features of focal pancreatic lesions in cats

“…The decision tree is a machine learning-based tool that has seldom been proposed in the veterinary medical literature for use by the clinician as a guide in interpreting both CEUS ( 19 ) and CT examinations ( 20 , 21 ). In the study by Burti et al ( 19 ), a decision tree was developed on 150 hepatic masses (used as a training set) and tested on another 35 cases (used as a test set).…”

“…It should be stated at this point that the classification accuracy for new cases could be lower than the accuracy resulting from cross-validation. The same cross-validation scheme has also been used in other studies ( 20 , 21 ). It is the authors' opinion that, despite the above-mentioned limitations, the proposed decision tree could act as a guide in classifying lesions based on their CEUS features.…”