Glucose is the primary source of energy for cells, which are the building blocks of life. It is given to the body by insulin that carries out the metabolic tasks that keep people alive. Glucose level imbalance is a sign of diabetes mellitus (DM), a common type of chronic disease. It leads to long-term complications, such as blindness, kidney failure, and heart disease, having a negative impact on one's quality of life. In Saudi Arabia, a ten-fold increase in diabetic cases has been documented within the last three years. DM is broadly categorized as Type 1 Diabetes (T1DM), Type 2 Diabetes (T2DM), and Pre-diabetes. The diagnosis of the correct type is sometimes ambiguous to medical professionals causing difficulties in managing the illness progression. Intensive efforts have been made to predict T2DM. However, there is a lack of studies focusing on accurately identifying T1DM and Pre-diabetes. Therefore, this study aims to utilize Machine Learning (ML) to distinguish and predict the three types of diabetes based on a Saudi Arabian hospital dataset to control their progression. Four different experiments have been conducted to achieve the highest results, where several algorithms were used, including Support Vector Machine (SVM), Random Forest (RF), K-Nearest Neighbor (K-NN), Decision Tree (DT), Bagging, and Stacking. In experiments 2, 3, and 4, the Synthetic Minority Oversampling Technique (SMOTE) was applied to balance the dataset. The empirical results demonstrated promising results of the novel Stacking model that combined Bagging K-NN, Bagging DT, and K-NN, with a K-NN meta-classifier attaining an accuracy, weighted recall, weighted precision, and cohen's kappa score of 94.48%, 94.48%, 94.70%, and 0.9172, respectively. Five principal features were identified to significantly affect the model accuracy using the permutation feature importance, namely Education, AntiDiab, Insulin, Nutrition, and Sex.

Background/Objectives: Hypoparathyroidism (HPT) is a common complication following thyroid surgery with an incidence reaching up to 29%, potentially resulting in significant long-term morbidity. To improve its early identification and patient outcomes, we investigated the prevalence and predictors of postoperative HPT. Methods: This retrospective, multicenter observational study included patients who underwent thyroid surgery from 2016 to 2022 in four centers located in Saudi Arabia’s Eastern Province. We analyzed demographic data, underlying thyroid or parathyroid conditions, surgical indications, types of procedures, pathology results, and preoperative corrected calcium and vitamin D levels, along with postoperative corrected calcium and parathyroid hormone (PTH) levels. For data analysis, IBM Statistical Package for the Social Sciences (SPSS) Statistics 22 was used, with categorical variables presented as frequencies/percentages and non-normal continuous variables as the median/first quartile (Q1) and third quartile (Q3). Associations were tested with chi-square/Fisher exact tests, medians with Mann–Whitney U-tests, and odds ratios (ORs) with 95% confidence intervals (CIs) via multivariate analysis with statistical significance set at p < 0.05. Results: A total of 679 cases were included. The median age of patients was 43 years (with 48.9% of them aged 41–60 years), and 82% were female. HPT occurred in 228 cases (35.3%), with 115 (81.0%) experiencing transient HPT and 27 (19.0%) permanent HPT. Multivariate analysis identified total thyroidectomy (OR 2.7, p = 0.005), completion thyroidectomy (OR 8.4, p = 0.004), and low immediate postoperative PTH level (OR 3.1, p < 0.001) as independent predictors of HPT. Central lymph node dissection (CLND; OR 4.03, p = 0.004) and low postoperative PTH level (OR 2.56, p = 0.049) were significant predictors of permanent HPT. Conclusions: Key predictors of HPT include surgical extent and low postoperative PTH level, while CLND and low postoperative PTH level are the strongest predictors of permanent HPT. Careful assessment of these risks when determining the extent of surgery and avoiding unnecessary aggressive procedures can help to minimize the occurrence of HPT. Measuring the PTH level immediately after surgery may aid in identifying high-risk patients for early intervention and appropriate follow-up.

Background and Objectives: Indeterminate thyroid nodules (Bethesda III and IV) are a common clinical entity that present a diagnostic challenge due to their intermediate risk of malignancy. This study aimed to evaluate the role of ultrasound in risk stratification and malignancy prediction to support clinical decision-making and reduce unnecessary surgical interventions. Materials and Methods: This retrospective multicenter cohort study included patients aged ≥18 years who underwent thyroid surgery between 2016 and 2022 at four centers in the Eastern Province of Saudi Arabia. Only nodules with indeterminate cytology (Bethesda III or IV) were included. Data collected included demographic characteristics, thyroid function, ultrasound features, cytology results, and histopathological findings. Results: A total of 679 patients with 733 nodules were reviewed. Of these, 206 patients with 223 indeterminate nodules were included (median age: 42 years; 88.3% female). The overall malignancy rate was 46.6%. Independent predictors of malignancy included solid hypoechoic composition (OR = 2.26, p = 0.012), microcalcifications (OR = 3.07, p = 0.002), lymph node involvement (OR = 2.43, p = 0.021), American Thyroid Association (ATA) intermediate to high suspicion category (OR = 1.9, p = 0.018), and Thyroid Imaging Reporting and Data Systems (TI-RADS) categories 4–5 (OR = 2.3, p = 0.003). Solid hypoechoic nodules showed 82.3% specificity and 63.0% positive predictive value (PPV); microcalcifications demonstrated 84.1% specificity and 68.4% PPV; lymph node involvement had 87.6% specificity and 68.9% PPV. The ATA and TI-RADS classifications showed higher sensitivity (63.5% and 68.0%, respectively), but lower specificity (53.1% and 52.8%, respectively). Conclusions: Ultrasound features, particularly solid hypoechoic composition, microcalcifications, and lymph node involvement, as well as ATA and TI-RADS classifications, were independent predictors of malignancy in indeterminate thyroid nodules. Although ATA and TI-RADS offered higher sensitivity, individual features demonstrated greater specificity and PPV. These findings support the use of ultrasound risk stratification to guide surgical decisions in high-risk cases and suggest that additional diagnostic evaluation may be appropriate for low-risk nodules.

Recognizing hepatic manifestations of COVID-19 and their impact on the severity and outcome is crucial in managing this emerging pandemic. However, we lack such reported data in Saudi Arabia regarding this clinical entity. This is a retrospective observational study conducted on 387 patients with COVID-19 disease who were hospitalized at King Fahad Hospital of the University from March-September 2020. The total cohort was divided into two groups: liver and non-liver involvement. Then, the frequency of hepatic manifestations was determined, followed by comparing severity and outcome among the two study groups. A total of 387 patients were included, of which 72.87% had hepatic manifestations. The most prevalent abnormalities were high LDH in 308 (79.58%) followed by AST 205 (52.97%), GGTP 124 (31.26%), ALT 74 (19.12%), PT/INR 66 (17.05%), direct bilirubin 51 (12.40%), total bilirubin 46 (11.88%), and low albumin 48 (12.4%). Univariate analyses showed that liver involvement was significantly associated with severe (31.91%) and critical (34.75%) presentation (P<0.001). Multivariate regression analysis showed that the presence of liver involvement was an independent risk factor for severe or critical COVID-19 disease (OR 2.44; P<0.001), longer hospitalization (OR 2.27; P=0.001), and ICU admission (OR 2.27; P=0.006). The current study showed that liver involvement is common in the setting of COVID-19 disease. Such patients had a higher disease severity and a worse clinical outcome.