Search Results (66)

Background: Fine-needle aspiration biopsy (FNAB) is the standard initial diagnostic procedure for thyroid nodules; however, a considerable proportion of results are non-diagnostic or indeterminate, often requiring repeat procedures and delaying management. Core needle biopsy (CNB) has been proposed as a second-line option. This study evaluated the frequency of non-conclusive FNAB and CNB results and assessed the diagnostic contribution of CNB in nodules with initially non-conclusive FNAB findings. Methods: A retrospective–prospective study was conducted between 2019 and 2025 at a tertiary referral center, including 434 thyroid nodules. Ultrasound risk stratification followed ACR TI-RADS criteria. FNAB was performed in 430 nodules, and CNB in 85 nodules, including 82 evaluated by both methods. Biopsy results were classified according to the Bethesda system as conclusive or non-conclusive. Paired comparisons were analyzed using the McNemar test, and associations with ultrasound risk were assessed. Results: FNAB produced non-conclusive results in 56.5% of cases, compared with 23.5% for CNB. In paired analysis, 53.7% of nodules with non-conclusive FNAB were reclassified as conclusive after CNB (p < 0.001). CNB significantly distinguished benign from malignant lesions, unlike FNAB. Hypoechogenicity, irregular margins, and punctate echogenic foci were independent predictors of malignancy. Minor complications were more frequent after CNB, while major complications were rare in both groups. Conclusions: CNB improves diagnostic yield when used as a second-line procedure in nodules with non-conclusive FNAB findings. Selective use in higher-risk nodules may reduce repeat procedures and facilitate more structured clinical management. Full article

Background/Objectives: This study aimed to automatically classify Zirconia-based fixed restorations and porcelain-fused-to-metal (PFM) restorations on panoramic radiographs using an artificial intelligence-based model. Unlike previous studies that mainly focused on classifying types of restorations (e.g., crowns, fillings, implants), this research concentrated on material-based differentiation, aiming to provide a more specific contribution to clinical decision support systems. Method: Panoramic radiographs obtained from the archive of Süleyman Demirel University Faculty of Dentistry were included in this study. Radiographs with poor image quality or insufficient visibility of the restoration area were excluded. A total of 593 cropped region-of-interest (ROI) images, labeled by expert prosthodontists using ImageJ software (version 1.54r; National Institutes of Health, Bethesda, MD, USA), were included in the analysis. In order to reduce class imbalance, data augmentation was applied only for images in the Zirconia-based fixed restorations class. By using various image processing techniques such as rotation, reflection and brightness change, the number of samples in the zirconia-based restorations class was increased and thus a balanced dataset was obtained with a close number of samples for both classes. For model training, the pre-trained VGG16 architecture was used with a transfer learning method, and the final layers were retrained and fine-tuned. The model was configured specifically for binary classification. The entire dataset was randomly split into 70% training, 20% validation, and 10% testing. Model performance was evaluated using accuracy, F1-score, sensitivity, and specificity. Results: The model correctly classified 90 out of 94 images in the test dataset, achieving an overall accuracy rate of 96%. For both classes, the precision, recall, and F1-score values were measured in the range of 95% to 96%. Additionally, the Area Under the Curve (AUC) of the ROC curve was calculated as 0.994, and the Average Precision (AP) score was determined to be 0.995. According to the confusion matrix results, only 4 images were misclassified, consisting of 2 false positives and 2 false negatives. Conclusions: The deep learning model demonstrated high accuracy in differentiating zirconia and metal-supported porcelain restorations on panoramic radiographs, suggesting that material-based AI classification may support clinical decision-making in restorative dentistry. Full article

Background: Risk stratification of indeterminate thyroid nodules (Bethesda III–IV) remains difficult and often triggers unnecessary procedures. Ultrasound-based ACR TI-RADS and the 2023 Bethesda System are widely used, but the incremental value of combining them and the role of size thresholds needs prospective validation. Objective: The objective of this study was to prospectively compare the diagnostic performance of ACR TI-RADS and the 2023 Bethesda System, alone and in combination, for predicting malignancy in thyroid nodules, with dedicated analyses of indeterminate lesions (Bethesda categories III–IV), including subtypes of Bethesda III (nuclear atypia vs. other atypia), and the impact of nodule size. Methods: Histopathology was available for 131 nodules. Diagnostic metrics (sensitivity, specificity, PPV, NPV), ROC curves (DeLong comparison), and Youden indices were calculated for individual and combined thresholds; a 16 mm size cut-off was explored. Results: Malignancy was confirmed in 105/131 nodules (80.2%). Bethesda outperformed TI-RADS (AUC 0.87 vs. 0.69; DeLong p = 0.041). Malignancy rates rose with higher categories (e.g., TI-RADS 5: 93.6%; Bethesda category V: 100%; Bethesda category VI: 100%) and were markedly elevated in the histologically confirmed subset for Bethesda category III (32/41; 78.0%) and IV (6/8; 75.0%). The combined requirement of TI-RADS ≥ 4 and Bethesda ≥ 4 maximized specificity (96.2%) and PPV (98.4%) with a high Youden J (0.552), supporting a rule-in strategy in category IV of Bethesda. Size alone was a weak discriminator (AUC 0.66); within Bethesda III–IV nodules, malignancy did not differ significantly by the 16 mm threshold (p = 1.00). ROC using continuous tumor size yielded AUC = 0.66; the ROC-derived optimal cut-off was 16 mm. Applying this split produced sensitivity 0.80 and specificity 0.50. Conclusions: Integrating ACR TI-RADS with Bethesda cytology significantly improves specificity and PPV for indeterminate thyroid nodules, supporting a morphology-driven approach over traditional size-based thresholds. Incorporation of combined sonographic–cytologic criteria into management algorithms may reduce unnecessary interventions and optimize patient care. Full article

Automated analysis of liquid-based cervical cytology is increasingly supported by digital microscopy and deep learning. However, model generalization remains challenging due to scanner- and laboratory-induced domain shifts affecting color, texture, and morphology. In this study, we present a robust cell-level classification framework for liquid-based Pap smear cytology based on deep transfer learning, designed to operate under heterogeneous acquisition conditions. We construct a multi-source dataset by integrating three widely used public reference repositories (SIPaKMeD, Herlev, CRIC Cervix) with a proprietary cohort comprising 416 Whole Slide Images (WSIs) collected from two medical centers and digitized using different scanning systems. All labels are harmonized into four Bethesda categories (NILM, ASC-US, LSIL, HSIL), and cell-centered 224 × 224 patches are used as standardized inputs for model development and benchmarking. We evaluate state-of-the-art CNN backbones (ResNet50, EfficientNetB0, VGG16) and perform systematic ablation across data-source combinations to quantify robustness under acquisition variability. Among the evaluated models, ResNet50 yields the best overall performance on the independent test set (accuracy = 0.91; macro-F1 = 0.91), consistently outperforming EfficientNetB0 and VGG16. Importantly, incorporating proprietary multi-center WSI-derived data improves robustness to scanner-induced variation compared to training on public data alone. These findings demonstrate that combining diverse data sources can mitigate domain shift in cell-level cervical cytology classification. While clinically actionable screening requires slide-level aggregation (e.g., MIL-based WSI inference), the proposed classifier provides a robust component that can be integrated into end-to-end WSI screening pipelines in future work. Full article

Background: Human papillomavirus (HPV) is the primary etiological agent of cervical cancer. Although obesity has been proposed as a factor influencing HPV acquisition and disease course through immune and metabolic mechanisms, its role remains controversial. This study aimed to evaluate the association between body mass index (BMI) and high-risk (HR)-HPV infection, including genotype distribution, infection type, and cytological findings. Methods: This cross-sectional study included women aged 21 years and older who underwent cervical sampling between August and November 2025. Participants were classified as non-obese (BMI <30 kg/m²) or obese (BMI ≥30 kg/m²). HR-HPV genotypes were detected using a multiplex real-time PCR method, and cytological evaluation was performed according to the Bethesda Cervical Cytology Reporting System. Results: Among 518 women, the overall HR-HPV positivity rate was 13.5%. No significant difference in HR-HPV positivity was observed between obese (11.6%) and non-obese (14.2%) women (OR = 0.79; 95% CI: 0.44–1.44; p = 0.452). After age adjustment, obesity was not identified as an independent risk factor for HR-HPV infection. BMI was not associated with HPV genotype distribution, infection type, or cytological findings (all p > 0.05). HPV-68 was the most frequently detected genotype. Conclusions: BMI was not independently associated with HR-HPV infection or related clinical and cytological features. These findings suggest that HPV infection is primarily influenced by viral characteristics and host immune response, while BMI appears to play a limited role. Further multicenter prospective studies are needed to clarify the impact of obesity on HPV infection. Full article

Background/Objectives: Human papillomavirus (HPV) infection remains the principal etiologic factor for cervical intraepithelial neoplasia (CIN) and cervical cancer. This longitudinal cohort study aimed to characterize the dynamics of cytological and histopathological changes over a two-year follow-up, focusing on post-treatment reduction in lesion grade, persistence, and progression in relation to HPV genotype distribution and smoking status. Methods: A total of 351 women aged 20–76 years were included, with cervical samples collected at the “Elena Doamna” Clinical Hospital, Iași, Romania. Cytology was categorized according to the Bethesda System, while colposcopy and conization served as diagnostic confirmation methods. HPV genotyping identified both high-risk (HR) and low-risk (LR) viral subtypes. Longitudinal assessments were performed at baseline, one-year, and two-year intervals to evaluate temporal patterns of disease evolution. Results: At baseline, HSIL represented the predominant cytologic category (51.3%, n = 180), followed by ASC-US (19.1%), ASC-H (15.1%), and LSIL (14.5%). Negative cytology increased from 62.4% at one year to 71.8% at two years, indicating substantial post-treatment reduction in lesion grade. Downgrading of lesion severity after treatment occurred in 26.2%, persistence in 11.1%, and progression in 11.1% of cases. Concordance between colposcopy and conization was moderate but statistically significant (κ = 0.345), with the highest agreement observed for HSIL with equivocal features between CIN II and CIN III lesions. Smoking showed a significant association with lesion persistence at two years (OR = 3.07; 95% CI: 1.16–8.08) but no statistically significant association with HR-HPV persistence. HR-HPV genotypes 16, 18, 31, and 33 were most frequently linked to progression, whereas HPV 35, 59, and 68 were associated with persistence. Conclusions: Over two years, most cervical lesions regressed or normalized, demonstrating effective management and follow-up. Persistent infection with HR-HPV types and smoking were the primary determinants of unfavorable outcomes. These findings highlight the clinical relevance of sustained surveillance, HPV genotyping, and smoking cessation as integral components of evidence-based cervical disease prevention and management strategies. Full article

Background/Objectives: Multimodal large language models (LLMs) may directly interpret medical images, including thyroid ultrasounds (USs). Whether these models can reliably assess thyroid nodules—where subtle echogenic and morphological details are critical—remains uncertain. The American College of Radiology (ACR) TI-RADS system provides a structured framework for benchmarking artificial intelligence. This study evaluates GPT-5.0’s ability to interpret thyroid US images according to TI-RADS criteria and contextualizes its performance relative to expert radiologist assessment, using FNA cytology as the reference standard. Methods: This retrospective study included 100 patients (mean age 49.8 ± 12.6 years; 72 women) with cytology-confirmed diagnoses: Bethesda II (benign) or Bethesda V–VI (malignant). Each nodule had longitudinal and transverse US images acquired with high-frequency linear probes. A board-certified radiologist (>10 years’ experience) and GPT-5.0 independently assessed TI-RADS features (composition, echogenicity, shape, margin, echogenic foci) and assigned final categories. Agreement was analyzed using Cohen’s κ, and diagnostic performance was calculated using TR4–TR5 as positive for malignancy. Results: Agreement was substantial for composition (κ = 0.62), shape (κ = 0.70), and margin (κ = 0.68); moderate for echogenicity (κ = 0.48); and poor for echogenic foci (κ = 0.12). GPT-5.0 demonstrated a systematic, risk-averse tendency to up-classify nodules, leading to increased TR4–TR5 assignments. Overall, the TI-RADS agreement was 58% (κ = 0.31). The radiologist showed superior diagnostic performance (sensitivity 89%, specificity 85%) compared with GPT-5.0 (sensitivity 67%, specificity 49%), largely driven by false-positive TR4 classifications among benign nodules. Conclusions: GPT-5.0 recognizes several high-level TI-RADS features but struggles with microcalcifications and tends to overestimate malignancy risk within a risk-stratification framework, limiting its standalone clinical use. Ultrasound-specific training and domain adaptation may enable meaningful adjunctive roles in thyroid nodule assessment. Full article

Background/Objectives: This study aimed to evaluate the relationship between preoperative clinical, ultrasonographic, and cytologic findings and postoperative histopathology in patients with thyroid nodules, and to determine diagnostic factors associated with malignancy. Materials and Methods: A retrospective analysis was conducted on 100 patients who underwent thyroid surgery between September 2012 and April 2014. Preoperative data—including clinical examination, thyroid function tests, and high-resolution ultrasonography—were compared with fine-needle aspiration biopsy (FNAB) results and final histopathology. Ultrasonographic features (echogenicity, calcification, vascularity, and margin) were analyzed for their association with malignancy. Statistical tests included chi-square, t-test, and correlation analysis (p < 0.05 considered significant). Results: Among 100 patients (79 females, 21 males; mean age 47.5 ± 13.9 years), 29 (29%) had benign and 71 (71%) malignant histopathology. Malignancy was significantly associated with older age (p = 0.025), smaller nodule size (p = 0.019), hypoechogenicity (p = 0.001), microcalcifications (p = 0.014), and irregular margins (p = 0.017). FNAB showed a strong correlation with final histopathology (r = 0.65, p = 0.001). The overall sensitivity and specificity of FNAB were 25.4% and 82.8%, respectively. Conclusions: Hypoechogenicity, microcalcifications, and irregular margins were the most reliable ultrasonographic predictors of malignancy. FNAB remains a highly specific but variably sensitive diagnostic tool, and its accuracy increases when interpreted in conjunction with ultrasonographic findings. Integrating cytology with structured imaging systems such as ACR TI-RADS and Bethesda classification enhances diagnostic precision in thyroid nodule evaluation. Full article

Background and Clinical Significance: DNA methylation profiling has revolutionized the classification of central nervous system (CNS) tumors, providing insights into tumor prognosis, recurrence, and personalized treatments. However, a significant challenge remains in classifying rare or molecularly undefined high-grade gliomas (HGGs) that fail to match existing methylation reference classes. This study evaluates the clinical, histopathological, and molecular characteristics of three unclassifiable cases through a retrospective analysis. Methylation profiling was performed by the National Institute of Health based on the 11b6 and 12b6 of the Heidelberg classifier, as well as the National Cancer Institute/Bethesda classifier. The cases were evaluated for histopathological features, molecular markers, and clinical outcomes. Case Presentation: We present three adult patients with histologically confirmed HGGs that were unclassifiable by standard DNA methylation profiling. All patients presented with diverse clinical and radiographic findings. Histopathological examination confirmed high-grade glial neoplasms in each case. However, methylation profiling failed to yield clear matches for any known class. Instead, profiling suggested indeterminate IDH-wildtype neoplasms with aggressive clinical courses. Following treatment, one patient experienced disease progression and died, while the other two remained without evidence of recurrence at follow-up. Conclusions: These cases underscore the persistent diagnostic challenges posed by CNS tumors that are unclassifiable by current DNA methylation, highlighting the urgent need for expanded reference datasets. While methylation profiling has transformed the field of tumor diagnostics, its limitations still exist. Enhanced collaboration to broaden diagnostic categories is essential to broaden diagnostic classifiers. Until these tools are refined, integration of clinical, histological, and molecular findings is imperative to optimize patient management, improve classification accuracy, and optimize therapeutic outcomes. Unclassifiable HGGs represent a critical gap in CNS tumor diagnostics. Addressing this requires global collaboration to enrich methylation databases. In the interim, a multimodal diagnostic strategy remains essential for the management of these challenging tumors. Full article

Background and Objectives: Thyroid nodules are common, and distinguishing benign from malignant lesions is essential for clinical decision-making. While EU-TIRADS provides ultrasound-based risk stratification, fine-needle aspiration biopsy (FNAB) and the Bethesda System remain central diagnostic tools. This study aimed to compare the diagnostic performance of EU-TIRADS and Bethesda classifications and to identify ultrasonographic features independently associated with malignancy. Materials and Methods: This retrospective single-center study included 824 patients (1132 nodules) who underwent FNAB between August 2021 and June 2024. All ultrasound examinations and FNAB procedures were performed by the same endocrinologist. Sonographic features, EU-TIRADS categories, Bethesda classes, surgical indications, and histopathology were analyzed. Diagnostic accuracy was assessed using ROC curves, and multivariable logistic regression was applied to determine independent predictors of malignancy. Results: Among all nodules, 51.0% were EU-TIRADS 3, 28.6% were EU-TIRADS 4, and 19.2% were EU-TIRADS 5. Bethesda class II constituted 62.7% of FNAB results. Of the 289 surgically treated nodules, 53.3% were malignant. Malignant nodules were smaller, more often solitary and unilateral, and more frequently located in the upper pole (p < 0.05). Irregular margins (OR = 8.15, p < 0.001) and microcalcifications (OR = 10.01, p = 0.003) were independent predictors of malignancy. Taller-than-wide shape also showed significant association. ROC analyses demonstrated that EU-TIRADS (AUC = 0.808) and Bethesda (AUC = 0.869) were both significant predictors, with Bethesda showing higher specificity. Malignancy rates were 0% in EU-TIRADS II, 4.3% in III, 14.5% in IV, and 37.8% in V. Conclusions: EU-TIRADS is a practical and sensitive non-invasive tool for malignancy risk stratification; however, Bethesda classification remains superior in overall diagnostic accuracy. Microcalcification and irregular margins were the strongest ultrasonographic predictors of malignancy, while macrocalcification, parenchymal heterogeneity, and thyroiditis showed no significant association. These findings support the complementary roles of EU-TIRADS and FNAB and highlight key sonographic markers that enhance malignancy prediction in thyroid nodule evaluation. Full article

Background/Objectives: Both benign and malignant thyroid lesions present as nodules. While thyroid nodules are less common in the pediatric population than in adults, their malignancy rates are considerably higher. Although the 10 mm cut-off for fine-needle aspiration cytology (FNAC) is commonly used for both adults and children, there is limited information regarding subcentimeter thyroid nodules in the pediatric population. The majority of published studies have focused on nodules measuring 1 cm or greater. This study aimed to compare the cytological diagnosis, ultrasonographic features, and histopathological outcomes of thyroid nodules in pediatric patients (under 21 years old), stratified by size (≤10 mm vs. >10 mm). Methods: We conducted a retrospective, single-center cohort study, evaluating 108 thyroid nodules from 98 patients. Nodule sizes were categorized into two groups, and their features were correlated with findings from FNAC using the Bethesda System for Reporting Thyroid Cytopathology and subsequent surgical histopathology. The risk of malignancy (ROM) was calculated for each Bethesda category. Results: A total of 108 nodules were evaluated, with 35 (32.4%) measuring ≤ 10 mm. The overall malignancy rate was 12%, with 14.3% in the ≤10 mm group and 11% in the >10 mm group. The difference was not statistically significant, and this finding indicates that small nodules can also harbor malignancy. Notably, all cases categorized as suspicious for malignancy or malignant by FNAC were confirmed to be malignant on histopathology (ROM = 100%). The Atypia of Undetermined Significance (AUS) category exhibited a malignancy rate of 60%, which is significantly higher than the rates reported in previous studies. Ultrasonographic features such as hypoechogenicity and microcalcifications were more prevalent in malignant nodules but lacked statistical significance. Conclusions: Our findings demonstrate that pediatric thyroid nodules, including those ≤10 mm, have a notable risk of malignancy. The high rate of malignancy in the AUS category suggests that the current Bethesda criteria, primarily designed for adults, may require re-evaluation for pediatric cases due to known differences in genetic profiles and disease behavior. Consequently, these pathological findings clearly demonstrate that FNAC indications in children should not be based solely on nodule size, and that a multidisciplinary approach guided by pediatric-specific guidelines should inform clinical management. Full article

Background: Microsurgical breast reconstruction is advancing rapidly with the integration of innovative technologies that enhance surgical precision, safety, and outcomes. This narrative review highlights recent developments across four key phases: flap planning, flap harvest, microvascular anastomosis, and flap monitoring. Methods: To identify the most updated and relevant data, all content on «Aesthetic and Reconstructive Breast Surgery Network» (ARBS Network, Copyright 2025 Mark Allen Group, United Kingdom) was screened regarding new technology. The contributions were grouped into one of four key phases. More references related to the content viewed were then searched on the electronic database MEDLINE (Bethesda, MD: U.S. National Library of Medicine). Results: 24 contributions regarding new technology were identified on ARBS Network. Of these, 17 were relevant for this paper. Preoperative tools such as CT angiography and AI-based perforator mapping optimize surgical planning and execution. Robotic-assisted or endoscopic techniques for deep inferior epigastric perforator (DIEP) flap harvest enable minimally invasive dissection with reduced donor-site morbidity and improved muscle preservation. Robotic microsurgery, particularly with the MUSA and Symani^® Surgical System, allows for precise, tremor-free suturing of submillimeter vessels. Indocyanine green (ICG) angiography remains the gold standard for intraoperative perfusion evaluation. Postoperative flap surveillance is crucial for detecting vascular compromise early. Devices such as the Cook-Swartz Doppler probe and flow couplers offer continuous monitoring. Wireless oximetry systems like ViOptix^® provide non-invasive, real-time perfusion data and support remote monitoring. Conclusions: Collectively, these innovations are transforming microsurgical breast reconstruction by increasing efficiency, consistency, and outcomes. Full article

Accurate preoperative assessment of thyroid nodules remains challenging, particularly in indeterminate cytological categories. Integrating molecular testing into cytology could improve diagnostic precision, enable timely intervention, and support better risk stratification and patient management. This proof-of-concept study evaluated the feasibility of performing molecular testing on fine-needle aspiration cytology (FNAC) samples processed on CytoMatrix, a three-dimensional synthetic scaffold designed to capture and preserve cellular material. Thirty-three thyroid FNAC specimens were processed on CytoMatrix, and cytological diagnoses were mirrored to the 2023 Bethesda System for Reporting Thyroid Cytopathology and correlated with final histopathology. DNA was extracted from paraffin-embedded CytoMatrix sections and analyzed for the BRAF V600E mutation. Adequate DNA for molecular testing was obtained in 30 of 33 cases (90%), and BRAF V600E mutations were detected in three papillary thyroid carcinoma samples. DNA adequacy and yield were consistent across Bethesda III–V categories, with insufficiency limited to low-cellularity Bethesda III cases. CytoMatrix enables reliable DNA recovery and targeted molecular testing without compromising cytological evaluation. This integrated cytomolecular workflow provides a feasible approach for combining cytological and molecular data in thyroid FNAC, supporting personalized and timely diagnostic management. Full article

Fine-needle aspiration cytology (FNAC) is the cornerstone of thyroid nodule evaluation, standardized by the Bethesda System. However, indeterminate categories (Bethesda III–IV) remain a major challenge, often leading to unnecessary surgery or delayed molecular testing. Deep learning (DL) has recently emerged as a promising adjunct in thyroid cytopathology, with applications spanning triage support, Bethesda category classification, and integration with molecular data. Yet, routine adoption is limited by preanalytical variability (staining, slide preparation, Z-stack acquisition, scanner heterogeneity), annotation bias, and domain shift, which reduce generalizability across centers. Most studies remain retrospective and single-institution, with limited external validation. This article provides a technical overview of DL in thyroid cytology, emphasizing preanalytical sources of variability, architectural choices, and potential clinical applications. We argue that standardized datasets, multicenter prospective trials, and robust explainability frameworks are essential prerequisites for safe clinical deployment. Looking forward, DL systems are most likely to enter practice as diagnostic co-pilots, Bethesda classifiers, and multimodal risk-stratification tools. With rigorous validation and ethical oversight, these technologies may augment cytopathologists, reduce interobserver variability, and help transform thyroid cytology into a more standardized and data-driven discipline. Full article

Background: Bethesda III and IV thyroid nodules, which fall under the category of indeterminate cytology, pose challenges in clinical decision-making due to their ambiguous risk of malignancy. Molecular testing has been increasingly employed to aid risk stratification and optimize the extent of surgical intervention. Methods: A retrospective review of 410 patients with Bethesda III and IV thyroid nodules who underwent thyroid surgery at McGill University teaching hospitals between January 2016 and April 2022. Patients were grouped based on whether or not they underwent preoperative molecular testing. Data were collected on demographic variables, histopathologic diagnosis, mutation profiles, and surgical outcomes. The primary outcome was to assess for concordance between surgical intervention and final pathology in both groups, with a focus on identifying optimal versus suboptimal management. Optimal management is defined as surgery appropriate to the aggressiveness of disease, meaning a hemi-thyroidectomy for a non-aggressive malignancy, total thyroidectomy for an aggressive malignancy, and no surgery for a benign nodule. Furthermore, suboptimal management includes unnecessary surgery or incorrect surgery for the level of aggressivity of the nodule. Results: Among the 410 patients, 203 underwent molecular testing, while 207 did not. Of those who underwent molecular testing, 117 had Bethesda III nodules and 86 had Bethesda IV nodules. In the non-tested group, 129 and 78 patients had Bethesda III and IV nodules, respectively. Optimal surgical intervention was achieved in 67.5% of patients who underwent molecular testing, compared with 25.1% in those who did not (p < 0.001). Subgroup analysis revealed that 61.5% of Bethesda III nodules with molecular testing received optimal care versus 21.0% of those without testing. In the Bethesda IV cohort, optimal surgery was achieved in 75.6% with testing versus 32.1% without. Among the suboptimally managed patients, 70.1% (155/221) were from the group that did not undergo molecular testing. In addition, molecular testing identified aggressive mutations such as BRAF V600E and TERT promoter mutations more frequently in Bethesda III nodules, while RAS-like mutations, associated with indolent behavior, predominated in Bethesda IV nodules. Conclusions: In this study, molecular testing significantly improved risk stratification and the likelihood of optimal surgical management in patients with Bethesda III and IV thyroid nodules. Incorporating molecular diagnostics into the standard preoperative workflow may enhance patient care, reduce unnecessary surgeries, and optimize the extent of surgery. Future studies should evaluate the cost-effectiveness and broader implementation of molecular testing in diverse healthcare settings. Full article