Machine Learning May Be An Alternative To BIPSS In The Differential Diagnosis Of ACTH-Dependent Cushing's Syndrome.

in The Journal of clinical endocrinology and metabolism by Ahmet Numan Demir, Deger Ayata, Ahmet Oz, Cem Sulu, Zehra Kara, Serdar Sahin, Dilan Ozaydin, Bora Korkmazer, Serdar Arslan, Osman Kizilkilic, Sema Ciftci, Ozlem Celik, Hande Mefkure Ozkaya, Necmettin Tanriover, Nurperi Gazioglu, Pinar Kadioglu

TLDR

  • This study developed a machine learning algorithm to help doctors diagnose a condition called Cushing's syndrome. The algorithm used information about a patient's biochemical and radiological features to determine if they had Cushing's disease or ectopic ACTH syndrome. The algorithm was very accurate, with a diagnostic accuracy of 86%. The most important factors that the algorithm used to make its diagnosis were the 2-day 2-mg dexamethasone suppression test, the > 50% suppression in the 8-mg high-dose dexamethasone test, late-night salivary cortisol, and the diameter of the pituitary adenoma. This study suggests that machine learning algorithms could be a useful tool in the diagnosis of Cushing's syndrome and could help doctors make more accurate diagnoses.

Abstract

This study aimed to develop machine learning (ML) algorithms for the differential diagnosis of adrenocorticotropic hormone (ACTH)-dependent Cushing's syndrome (CS) based on biochemical and radiological features. Logistic regression algorithms were used for ML, and the area under the receiver operating characteristics curve (AUROC) was used to measure performance. We used Shapley Contributed Comments (SHAP) values, which help explain the results of the ML models to identify the meaning of each feature and facilitate interpretation. A total of 106 patients, 80 with Cushing's disease (CD) and 26 with ectopic ACTH syndrome (EAS), were enrolled in the study. The ML task was created to classify patients with ACTH-dependent CS into CD and EAS. The average AUROC value obtained in the cross-validation of the logistic regression model created for the classification task was 0.850. The diagnostic accuracy of the algorithm was 86%. The SHAP values indicated that the most important determinants for the model were the 2-day 2-mg dexamethasone suppression test, the > 50% suppression in the 8-mg high-dose dexamethasone test, late-night salivary cortisol, and the diameter of the pituitary adenoma. We have also made our algorithm available to all clinicians via a user-friendly interface. ML algorithms have the potential to serve as an alternative decision support tool to invasive procedures in the differential diagnosis of ACTH-dependent CS.

Overview

  • The study aimed to develop machine learning algorithms for the differential diagnosis of adrenocorticotropic hormone (ACTH)-dependent Cushing's syndrome (CS) based on biochemical and radiological features. Logistic regression algorithms were used for ML, and the area under the receiver operating characteristics curve (AUROC) was used to measure performance. The study used Shapley Contributed Comments (SHAP) values to explain the results of the ML models and identify the meaning of each feature and facilitate interpretation. A total of 106 patients, 80 with Cushing's disease (CD) and 26 with ectopic ACTH syndrome (EAS), were enrolled in the study. The ML task was created to classify patients with ACTH-dependent CS into CD and EAS. The average AUROC value obtained in the cross-validation of the logistic regression model created for the classification task was 0.850. The diagnostic accuracy of the algorithm was 86%.

Comparative Analysis & Findings

  • The study compared the outcomes observed under different experimental conditions or interventions, specifically the use of logistic regression algorithms for the differential diagnosis of ACTH-dependent CS. The results showed that the logistic regression model created for the classification task had an average AUROC value of 0.850 and a diagnostic accuracy of 86%. The SHAP values indicated that the most important determinants for the model were the 2-day 2-mg dexamethasone suppression test, the > 50% suppression in the 8-mg high-dose dexamethasone test, late-night salivary cortisol, and the diameter of the pituitary adenoma. These findings suggest that logistic regression algorithms can be a useful tool in the differential diagnosis of ACTH-dependent CS.

Implications and Future Directions

  • The study's findings have significant implications for the field of research and clinical practice, as they suggest that logistic regression algorithms can be a useful tool in the differential diagnosis of ACTH-dependent CS. However, the study also identified some limitations, such as the need for more data to validate the results and the potential for overfitting. Future research could address these limitations by collecting more data, using different ML algorithms, and incorporating additional features. The study also suggests that logistic regression algorithms could be used as an alternative decision support tool to invasive procedures in the differential diagnosis of ACTH-dependent CS, which could improve patient outcomes and reduce healthcare costs.
Read Full Article