Diagnostic models for predicting malignancy in thyroid nodules classified as Bethesda Category III in an endemic region.

BACKGROUND: This study aims to develop a diagnostic model to help physicians determine whether thyroid nodules categorized as atypia of undetermined significance (AUS) in category III of the Bethesda system are benign or malignant preoperatively. To create a diagnostic model for predicting thyroid nodules' benign or malignant with AUS cytology based on clinical, ultrasonographic, and cytopathological findings. METHODS: This is a retrospective cohort study involving patients (>19) at risk of thyroid cancer who had thyroidectomy after an AUS cytology. The dataset consists of 53 variables 204 nodules from 183 patients. Binary logistic regression and factor analysis methods were used to identify risk factors for malignancy. Finally, four prediction models were developed using different approaches, based on clinical, pathological clinical + pathological, and the factors. RESULTS: A total of 88 (48.1%) of 183 patients diagnosed with AUS were benign and 95 (51.9%) the malignant. After determining risk factors, four prediction models were developed based on different approaches to assist physicians in deciding to detect AUS nodules early. It was seen that bilaterality was found to be a risk factor for malignancy in the clinical model (pbilaterality = .03) and it was also seen that the pathological variables pale chromatin and irregular contours in the oncocyte variables were risk factors for malignancy (ppalechromatin = .02, pirregularcontoursintheoncocyte = .04). The best model obtained sensitivity and specificity values are 73% and 87% based on clinical and pathological variables. CONCLUSION: This comprehensive study may provide a more in-depth understanding of AUS and make a notable contribution to healthcare professionals before surgery.

A novel computer based risk prediction model for vocal cord palsy before thyroidectomy.

BACKGROUND: Thyroidectomy may be performed for clinical indications that include malignancy, benign nodules or cysts suspicious findings on fine needle aspiration (FNA) biopsy, dyspnea from airway compression or dysphagia from cervical esophageal compression, etc. The incidences of vocal cord palsy (VCP) caused by thyroid surgery were reported to range from 3.4% to 7.2% and 0.2% to 0.9% for temporary and permanent vocal fold palsy respectively which is a serious complication of thyroidectomy that is worrisome for patients. OBJECTIVE: Therefore, it is aimed to determine the patients who have the risk of developing vocal cord palsy before thyroidectomy by using machine learning methods in the study. In this way, the possibility of developing palsy can be reduced by applying appropriate surgical techniques to individuals in the high-risk group. METHOD: For this aim, 1039 patients with thyroidectomy, between the years 2015 and 2018, have been used from Karadeniz Technical University Medical Faculty Farabi Hospital at the department of general surgery. The clinical risk prediction model was developed using the proposed sampling and random forest classification method on the dataset. CONCLUSION: As a result, a novel quite a satisfactory prediction model with 100% accuracy was developed for VCP before thyroidectomy. Using this clinical risk prediction model, physicians can be helped to identify patients at high risk of developing post-operative palsy before the operation.

Prediction of gestational diabetes using deep learning and Bayesian optimization and traditional machine learning techniques.

The study aimed to develop a clinical diagnosis system to identify patients in the GD risk group and reduce unnecessary oral glucose tolerance test (OGTT) applications for pregnant women who are not in the GD risk group using deep learning algorithms. With this aim, a prospective study was designed and the data was taken from 489 patients between the years 2019 and 2021, and informed consent was obtained. The clinical decision support system for the diagnosis of GD was developed using the generated dataset with deep learning algorithms and Bayesian optimization. As a result, a novel successful decision support model was developed using RNN-LSTM with Bayesian optimization that gave 95% sensitivity and 99% specificity on the dataset for the diagnosis of patients in the GD risk group by obtaining 98% AUC (95% CI (0.95-1.00) and p < 0.001). Thus, with the clinical diagnosis system developed to assist physicians, it is planned to save both cost and time, and reduce possible adverse effects by preventing unnecessary OGTT for patients who are not in the GD risk group.

A prospective study for gestational diabetes mellitus: Analysis of risk factors in Turkish women for early prediction.

OBJECTIVE: To define risk factors for the early prediction of gestational diabetes mellitus (GDM) because the risk of pre-eclampsia and preterm birth increases in mothers who are diagnosed with GDM. MATERIALS AND METHODS: A prospective study was designed and the data were collected by physicians prospectively from the patients who came to the clinic between the years 2019 and 2021; informed consent was obtained from the women. The prospective data comprised 489 patient records with 72 variables and the risk factors for early prediction of GDM were determined using logistic regression and random forest (RF), which is an advanced analysis method. RESULTS: The obtained sensitivity and specificity values are 90% and 75% for logistic regression and 71% and 90% for the RF, respectively. CONCLUSION: In this prospective study of GDM in Turkish women; age, body mass index, level of hemoglobin A1c, level of fasting blood sugar, physical activity time in first trimester, gravidity, triglycerides, and high-density lipoprotein cholesterol were confirmed to be risk factors in analysis results.

Comparison of enhancement methods for mammograms with performance measures.

Mammograms are generally contaminated by noise which assures the need for image enhancement to aid interpretation. The enhancement of mammograms is a very important problem for easy extraction of suspicious regions known as regions of interest (ROIs). This paper introduces comparison of various hybrid enhancement algorithms based on mathematical morphology, contrast stretching, wavelet transform, anisotropic diffusion filter and contrast limited adaptive histogram equalization (CLAHE). The performances of algorithms have been compared by using three global image enhancement evaluation measures; Enhancement Measure (EME), Absolute Mean Brightness Error (AMBE) and Peak Signal-to-Noise Ratio (PSNR). For this study, we have used MIAS database. Experimental results show that the combination of mathematical morphology, anisotropic diffusion filter and CLAHE methods, yields significantly superior image quality and provides more visibility for the suspicious regions.

A novel automatic suspicious mass regions identification using Havrda & Charvat entropy and Otsu's N thresholding.

Mass detection is a very important process for breast cancer diagnosis and computer aided systems. It can be very complex when the mass is small or invisible because of dense breast tissue. Therefore, the extraction of suspicious mass region can be very challenging. This paper proposes a novel segmentation algorithm to identify mass candidate regions in mammograms. The proposed system includes three parts: breast region and pectoral muscle segmentation, image enhancement and suspicious mass regions identification. The first two parts have been examined in previous studies. In this study, we focused on suspicious mass regions identification using a combination of Havrda & Charvat entropy method and Otsu's N thresholding method. An open access Mammographic Image Analysis Society (MIAS) database, which contains 59 masses, was used for the study. The proposed system obtained a 93% sensitivity rate for suspicious mass regions identification in 56 abnormal and 40 normal images.