Role of machine learning algorithms in predicting the treatment outcome of uterine fibroids using high-intensity focused ultrasound ablation with an immediate nonperfused volume ratio of at least 90.

OBJECTIVE: This study aimed to investigate the role of machine learning (ML) classifiers to determine the most informative multiparametric (mp) magnetic resonance imaging (MRI) features in predicting the treatment outcome of high-intensity focused ultrasound (HIFU) ablation with an immediate nonperfused volume (NPV) ratio of at least 90%. PATIENTS AND METHODS: Seventy-three women who underwent HIFU treatment were divided into groups A (n=47) and B (n=26), comprising patients with an NPV ratio of at least 90% and <90%, respectively. An ensemble feature ranking model was introduced based on the score values assigned to the features by five different ML classifiers to determine the most informative mpMRI features. The relationship between the mpMRI features and the immediate NPV ratio of 90% was evaluated using Pearson's correlation coefficients. The diagnostic ability of the ML classifiers was evaluated using standard performance metrics, including the area under the receiver operating characteristic curve, accuracy, sensitivity, and specificity in eight folds cross-validation. RESULTS: For all the 12 most informative features, the area under receiver operating characteristic curve (AUROC), accuracy, specificity, and sensitivity ranged from 0.5 to 0.97, 0.34 to 0.97, 0.56 to 1.0, and 0.87 to 1.0, respectively. The gradient boosting (GBM) classifier demonstrated the best predictive performance with an AUROC of 0.95 and accuracy of 0.92, followed by the random forest, AdaBoost, logistic regression, and support vector classifiers, which yielded an AUROC of 0.92, 0.92, 0.83, and 0.78 and accuracy of 0.96, 0.88, 0.84, and 0.84, respectively. GBM had the best classifier performance with the best performing features from each mpMRI group, Ktrans ratio of the fibroid to the myometrium, the ratio of area under the curve of the fibroid to the myometrium, subcutaneous fat thickness, the ratio of apparent diffusion coefficient value of fibroid to the myometrium, and T2-signal intensity of the fibroid. CONCLUSIONS: The preliminary findings of this study suggest that the most informative and best performing features from each mpMRI group should be considered for predicting the treatment outcome of HIFU ablation to achieve an immediate NPV ratio of 90%.

TWIST1 Gene Expression as a Biomarker for Predicting Primary Doxorubicin Resistance in Breast Cancer.

Doxorubicin is one of the most commonly used chemotherapeutic agents for adjuvant chemotherapy of breast cancer. In the studies focused on finding biomarkers to predict the response of the patients and tumors to the drugs used, the Twist transcription factor has been suggested as a candidate biomarker for predicting chemo-resistance of breast tumors. In this study, we aimed to investigate the relationship between TWIST transcription factor expression and the effectiveness of doxorubicin treatment on directly taken primary tumor samples from chemotherapy-naive breast cancer patients. Twenty-six primary breast tumor samples taken from 26 different breast cancer patients were included in this study. Adenosine triphosphate tumor chemo-sensitivity assay (ATP-TCA) has been used to determine tumor response to doxorubicin and real-time reverse-transcription polymerase chain reaction (RT-PCR) was used for analyzing the TWIST1 gene expression of tumors. There was a significant difference in TWIST gene expression between responder and non responder tumors (p <0.05). The TWIST gene expression of the drug-resistant group was higher than the responsive group. This difference was not dependent on the histopathological features of tumors. In conclusion, compatible with earlier studies that have been performed with cell lines, the current study supports the role of higher TWIST gene expression as a biomarker for predicting the response of breast tumors to chemo-therapeutic agent doxorubicin.