Search Results (15)

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: Reliable and accurate prediction of treatment response to neoadjuvant chemoradiotherapy (nCRT) in locally advanced rectal cancer (LARC) is usually demanding and continues to pose a challenge. Kurtosis as a histogram parameter calculated on T2-weighted MRI sequences might be an additional tool, as it represents a quantitative biomarker for response prediction. It is defined as a measure of distributions’ tails relative to the center of the distribution curve, which reflects tissue heterogeneity. The aim of the study was to evaluate the added value of T2-weighted kurtosis in predicting pathological response to nCRT in patients with LARC. Methods: a single-center cohort study included 71 patients with LARC who underwent both initial and post-nCRT MRI examinations followed by surgical resection in the form of the total mesorectal excision (TME). Histogram analysis was performed using software MIPAV (Medical Image Processing, Analysis, and Visualization, version 11.3.2, developed by the National Institutes of Health, Bethesda, MD, USA) on T2-weighted sequences, extracting kurtosis along with other histogram parameters. Pathological tumor regression grade (pTRG) in accordance with Mandard classification was considered the gold standard. Patients were classified as responders (pTRG 1–2) or non-responders (pTRG 3–5). Results: while other histogram parameters did not show statistically significant differences between groups, post-treatment values of kurtosis were significantly higher in responders compared to non-responders (4.28 ± 0.73 vs. 3.01 ± 0.17, p = 0.024). The F1 score as a classification metric (0.821) indicates an improvement in classification performance following therapy. Conclusions: T2-weighted kurtosis might be a significant tool in predicting pathological response to nCRT, representing a potentially valuable quantitative biomarker that could improve treatment response assessment. Full article

Background and Objectives: To identify clinical, ultrasound (US) and real-time elastography (RTE) characteristics indicative of malignancy in small thyroid lesions. Materials and Methods: 141 consecutive patients with incidentally discovered solid thyroid nodules (diameter ≤ 10 mm) by neck US were assessed, and RTE was performed. The nodules were classified per American (ACR-TIRADS) and European (EU-TIRADS) criteria; US-guided FNA was conducted on EU-TIRADS 5 nodules. The US and RTE features of nodules classified as benign (Bethesda II) or malignant (Bethesda V and VI) were compared. Results: 41 nodules were classified as EU-TIRADS 5. Their Fine Needle Aspiration (FNA) cytology was Bethesda II (n = 11), III-IV (n = 3), V (n = 10) or VI (n = 17). Bethesda V–VI patients had a higher rate of autoimmune thyroiditis (p = 0.015) and higher ACR-scoring points (p < 0.001) compared with Bethesda II. The elastography ratio was equal between the groups (p = 0.584). In logistic regression analysis, ACR-scoring points predicted FNA results, with an area under the curve (AUC) of 0.993 (sensitivity 92.6% and specificity of 100%). The clinical model (age, body mass index, sex, autoimmunity, L-thyroxine treatment, nodule diameter, elastography ratio) achieved an AUC of 0.744. A “mixed” model, combining clinical characteristics with the ACR scoring points, achieved perfect performance (AUC = 1.000), predicting FNA results with 100% sensitivity and specificity. Conclusions: The proposed “mixed model” can predict Bethesda V–VI in thyroid nodules <10 mm, allowing for the selection of those needing further evaluation. Full article

Background/Objectives: Accurate diagnosis of thyroid cancer is critical but challenging due to overlapping ultrasound (US) features of benign and malignant nodules. This study aimed to evaluate the diagnostic performance of non-invasive and minimally invasive US techniques, including B-mode US, shear wave elastography (SWE), color Doppler, superb microvascular imaging (SMI), and TI-RADS, in patients with suspected thyroid lesions and to assess their reliability and cost effectiveness compared with fine needle aspiration (FNA) biopsy. Methods: A prospective, single-center study (October 2023–February 2025) enrolled 300 patients with suspected thyroid cancer at a Spanish tertiary hospital. Of these, 296 patients with confirmed diagnoses underwent B-mode US, SWE, Doppler, SMI, and TI-RADS scoring, followed by US-guided FNA and Bethesda System cytopathology. Lasso-penalized logistic regression and a bootstrap analysis (1000 replicates) were used to develop diagnostic models. A utility function was used to balance diagnostic reliability and cost. Results: Thyroid cancer was diagnosed in 25 patients (8.3%). Elastography combined with SMI achieved the highest diagnostic performance (Youden index: 0.69; NPV: 97.4%; PPV: 69.1%), outperforming Doppler-only models. Intranodular vascularization was a significant risk factor, while peripheral vascularization was protective. The utility function showed that, when prioritizing cost, elastography plus SMI was cost effective (α < 0.716) compared with FNA. Conclusions: Elastography plus SMI offers a reliable, cost-effective diagnostic rule for thyroid cancer. The utility function aids clinicians in balancing reliability and cost. SMI and generalizability need to be validated in higher prevalence settings. Full article

The prevalence of goiter, thyroid nodules, and thyroid cancers in the pediatric population has increased. In some rare cases, local conditions such as juvenile desmoid-type fibromatosis (JDTF) can mimic specific thyroid pathology, complicating the diagnostic process. A 17-year-old obese adolescent girl was admitted to the Endocrinology Department with progressive swelling on the left side of the neck, persisting for approximately one year, recently accompanied by dysphonia and inspiratory dyspnea, and ultimately diagnosed as a unilateral nodular goiter associated with compressive phenomena. Despite her euthyroid status, the thyroid ultrasound identified a suspected, large, non-homogeneous, hypoechogenic nodule with calcifications in the left thyroid lobe (TI-RADS score of 4), confirmed by a cervical-region MRI. The biopsy specimens obtained through fine-needle aspiration were classified as Bethesda III (“atypia of undetermined significance” or “follicular lesion of undetermined significance”). Left thyroid lobe removal was performed by a specialized surgeon in thyroid pathology, with histopathological analysis revealing a diagnosis of JDTF in the thyroid gland. Post-surgery, the patient showed favorable progress without any relapse. Pediatric endocrinologists face challenges in diagnosing and managing thyroid nodules in children due to their higher malignancy potential. Familiarity with similar conditions, such as JDTF, is crucial in accurate diagnosis and appropriate pediatric management. Full article

The physiological impact of walking in nature was quantified via continuous heart rate variability (HRV), pre- and post-walk saliva cortisol measures, and self-reported mood and mindfulness scores for N = 17 participants who walked “The Green Road” at Walter Reed National Military Medical Center in Bethesda, Maryland. For N = 15 of the participants, HRV analysis revealed two main groups: group one individuals had a 104% increase (mean) in the root mean square standard deviation (RMSSD) and a 47% increase (mean) in the standard deviation of NN values (SDNN), indicating an overall reduction in physiological stress from walking the Green Road, and group two individuals had a decrease (mean) of 42% and 31% in these respective HRV metrics, signaling an increase in physiological stresses. Post-walk self-reported scores for vigor and mood disturbance were more robust for the Green Road than for a comparable urban road corridor and showed that a higher HRV during the walk was associated with improved overall mood. Saliva cortisol was lower after taking a walk for all participants, and it showed that walking the Green Road elicited a significantly larger reduction in cortisol of 53%, on average, when compared with 37% of walking along an urban road. It was also observed that the order in which individuals walked the Green Road and urban road also impacted their cortisol responses, with those walking the urban road before the Green Road showing a substantial reduction in cortisol, suggesting a possible attenuation effect of walking the Green Road first. These findings provide quantitative data demonstrating the stress-reducing effects of being in nature, thus supporting the health benefit value of providing access to nature more broadly in many settings. Full article

Introduction: Thyroid nodules are extremely common and require complex management to prevent unnecessary surgical intervention and ensure that no malignant disease is overlooked. Several diagnostic tools and scoring systems are available to evaluate the risk of malignancy (ROM). The goal is to assess variables that can aid and support the clinical recommendations suggested by the updated Bethesda System for Reporting Thyroid Cytopathology (TBSRTC-2023), such as the ultrasonographic features of thyroid nodules, particularly for the indeterminate categories III (atypia of undetermined significance) and IV (follicular neoplasm). Methods: We retrospectively analysed the correlation of the demographic and ultrasonographic characteristics of thyroid nodules with the cytopathological and histopathological diagnoses of TBSRTC categories III (atypia of undetermined significance), IV (follicular neoplasm), V (suspicious for malignancy), and VI (malignant) in patients who underwent surgery in a single Portuguese centre over a 10-year period. Results: In total, 360 nodules were evaluated in 341 patients, and 57% were histopathologically malignant or borderline. The majority were included in the TBSRTC indeterminate categories III and IV, with ROMs of 44% and 43%, respectively. The ultrasonographic characteristics associated with a higher TBSRTC category and a greater ROM value were hypoechogenicity, the presence of microcalcifications, irregular margins, and the presence of cervical adenopathy. When correlating with a malignant histology, only adenopathy and the presence of microcalcifications were observed to be statistically significant. Discussion: The indeterminate categories of the TBSRTC have been the most challenging ones to manage. The new TBSRTC (2023) guidelines, as well as the ultrasonographic characteristics of a patient’s nodule, can be helpful in assessing the ROM and deciding on an appropriate course of treatment. Other resources, such as molecular tests, are also playing a more important role in the clinical decision process and may become crucial in the future. Conclusions: The worrisome ultrasound features that this study found to statistically correlate with a malignant histology were the presence of microcalcifications and adenopathy. The clinical management of thyroid nodules requires a careful analysis of clinical history and an evaluation of demographic details, personal and family history, ultrasonographic features, and the results of cytopathology, thyroid function, and molecular/genetic tests. Full article

Objective: In clinical practice, thyroid nodules are typically visually evaluated by expert physicians using 2D ultrasound images. Based on their assessment, a fine needle aspiration (FNA) may be recommended. However, visually classifying thyroid nodules from ultrasound images may lead to unnecessary fine needle aspirations for patients. The aim of this study is to develop an automatic thyroid ultrasound image classification system to prevent unnecessary FNAs. Methods: An automatic computer-aided artificial intelligence system is proposed for classifying thyroid nodules using a fine-tuned deep learning model based on the DenseNet architecture, which incorporates an attention module. The dataset comprises 591 thyroid nodule images categorized based on the Bethesda score. Thyroid nodules are classified as either requiring FNA or not. The challenges encountered in this task include managing variability in image quality, addressing the presence of artifacts in ultrasound image datasets, tackling class imbalance, and ensuring model interpretability. We employed techniques such as data augmentation, class weighting, and gradient-weighted class activation maps (Grad-CAM) to enhance model performance and provide insights into decision making. Results: Our approach achieved excellent results with an average accuracy of 0.94, F1-score of 0.93, and sensitivity of 0.96. The use of Grad-CAM gives insights on the decision making and then reinforce the reliability of the binary classification for the end-user perspective. Conclusions: We propose a deep learning architecture that effectively classifies thyroid nodules as requiring FNA or not from ultrasound images. Despite challenges related to image variability, class imbalance, and interpretability, our method demonstrated a high classification accuracy with minimal false negatives, showing its potential to reduce unnecessary FNAs in clinical settings. Full article

Introduction: Thyroid nodule incidence is increasing due to the widespread application of ultrasonography. Fine-needle aspiration cytology is widely applied for the detection of malignancies. The aim of this study was to evaluate the predictive value of ultrasonography in thyroid cancer. Methods: This retrospective study included patients that underwent total thyroidectomy for benign thyroid disease or well-differentiated thyroid carcinoma from January 2017 to December 2022. The study population was divided into groups: the well-differentiated thyroid cancer group and the control group with benign histopathological reports. Results: In total, 192 patients were enrolled in our study; 159 patients were included in the well-differentiated thyroid cancer group and 33 patients in the control group. Statistical analysis demonstrated that ultrasonographic findings such as microcalcifications (90.4%), hypoechogenicity (89.3%), irregular margins (92.2%) and taller-than-wide shape (90.5%) were correlated to malignancy (p < 0.001). Uni- and multivariate analysis revealed that both US score (OR: 2.177; p < 0.001) and Bethesda System (OR: 1.875; p = 0.002) could predict malignancies. In terms of diagnostic accuracy, the US score displayed higher sensitivity (64.2% vs. 33.3%) and better negative predictive value (34.5% vs. 24.4%) than the Bethesda score, while both scoring systems displayed comparable specificities (90.9% vs. 100%) and positive predictive values (97.1% vs. 100%). Discussion: The malignant potential of thyroid nodules is a crucial subject, leading the decision for surgery. Ultrasonography and fine-needle aspiration cytology are pivotal examinations in the diagnostic process, with ultrasonography demonstrating better negative predictive value. Full article

Liquid-based cytology (LBC) plays a crucial role in the effective early detection of cervical cancer, contributing to substantially decreasing mortality rates. However, the visual examination of microscopic slides is a challenging, time-consuming, and ambiguous task. Shortages of specialized staff and equipment are increasing the interest in developing artificial intelligence (AI)-powered portable solutions to support screening programs. This paper presents a novel approach based on a RetinaNet model with a ResNet50 backbone to detect the nuclei of cervical lesions on mobile-acquired microscopic images of cytology samples, stratifying the lesions according to The Bethesda System (TBS) guidelines. This work was supported by a new dataset of images from LBC samples digitalized with a portable smartphone-based microscope, encompassing nucleus annotations of 31,698 normal squamous cells and 1395 lesions. Several experiments were conducted to optimize the model’s detection performance, namely hyperparameter tuning, transfer learning, detected class adjustments, and per-class score threshold optimization. The proposed nucleus-based methodology improved the best baseline reported in the literature for detecting cervical lesions on microscopic images exclusively acquired with mobile devices coupled to the µSmartScope prototype, with per-class average precision, recall, and F1 scores up to 17.6%, 22.9%, and 16.0%, respectively. Performance improvements were obtained by transferring knowledge from networks pre-trained on a smaller dataset closer to the target application domain, as well as including normal squamous nuclei as a class detected by the model. Per-class tuning of the score threshold also allowed us to obtain a model more suitable to support screening procedures, achieving F1 score improvements in most TBS classes. While further improvements are still required to use the proposed approach in a clinical context, this work reinforces the potential of using AI-powered mobile-based solutions to support cervical cancer screening. Such solutions can significantly impact screening programs worldwide, particularly in areas with limited access and restricted healthcare resources. Full article

Background: Molecular testing has been used as an adjunct to morphological evaluation in the workup of thyroid nodules. This study investigated the impact of two gene fusions, RET/PTC and THADA/IGF2BP3, that have been described as oncogenic events in thyroid neoplasms. Methods: We performed a retrospective, single-centered study at a McGill University teaching hospital in Montreal, Canada, from January 2016 to August 2021. We included patients who underwent surgery for thyroid nodules that pre-operatively underwent molecular testing showing either RET/PTC or THADA/IGF2BP3 gene fusion. Results: This study included 697 consecutive operated thyroid nodules assessed using molecular testing, of which five had the RET/PTC fusion and seven had the THADA/IGF2BP3 fusion. Of the five nodules in the RET/PTC group, 100% were malignant and presented as Bethesda V/VI. Eighty percent (4/5) were found to have lymph node metastasis. Twenty percent (1/5) had extrathyroidal extensions. Sixty percent (3/5) were a diffuse sclerosing variant of papillary thyroid carcinoma, and the rest were the classical variant. Of the seven THADA/IGF2BP3 nodules, all presented as Bethesda III/IV and 71.4% (5/7) were malignant based on the final pathology analysis, and 28.6% (2/7) were NIFTP. All the THADA/IGF2BP3 fusion malignancies were a follicular variant of papillary thyroid carcinoma. None had lymph node metastasis or displayed extrathyroidal extensions. Conclusions: RET/PTC nodules presented as Bethesda V/VI and potentially had more aggressive features, whereas THADA/IGF2BP3 nodules presented as Bethesda III/IV and had more indolent behavior. This understanding may allow clinicians to develop more targeted treatment plans, such as the extent of surgery and adjuvant radioactive iodine treatment. Full article

The risk of malignancy in thyroid nodules correlates with the presence of ultrasonographic features. In adults, ultrasound risk-classification systems have been proposed to indicate the need for further invasive diagnosis. Furthermore, elastography has been shown to support differential diagnosis of thyroid nodules. The purpose of our study was to assess the application of the American Thyroid Association (ATA), British Thyroid Association (BTA) ultrasound risk-classification systems and strain elastography in the management of thyroid nodules in children and adolescents from one center. Seventeen nodules with Bethesda III, IV, V and VI were selected from 165 focal lesions in children. All patients underwent ultrasonography and elastography followed by fine needle aspiration biopsy. Ultrasonographic features according to the ATA and BTA stratification systems were assessed retrospectively. The strain ratio in the group of thyroid nodules diagnosed as malignant was significantly higher than in benign nodules (6.07 vs. 3.09, p = 0.036). According to the ATA guidelines, 100% of malignant nodules were classified as high suspicion and 73% of benign nodules were assessed as low suspicion. Using the BTA U-score classification, 80% of malignant nodules were classified as cancerous (U5) and 20% as suspicious for malignancy (U4). Among benign nodules, 82% were classified as indeterminate or equivocal (U3) and 9% as benign (U2). Our results suggest that application of the ATA or BTA stratification system and elastography may be a suitable method for assessing the level of suspected malignancy in thyroid nodules in children and help make a clinical decision about the need for further invasive diagnosis of thyroid nodules in children. Full article

Background and Objectives: The aim of the study was to evaluate the predictive value of the ultrasound criterion “non-marked hypoechogenicity” for malignancy and to determine whether classification of these nodules as TIRADS 3 could improve the overall accuracy of consequently adjusted M-TIRADS score. Materials and Methods: A total of 767 patients with 795 thyroid nodules were subject to ultrasonography examination and ultrasound-guided fine needle aspiration biopsy. Nodules were classified by Kwak TIRADS and modified (M-TIRADS) categories 4A, 4B, and 5 according to number of suspicious US features (marked hypoechogenicity, microlobulated or irregular margins, microcalcifications, taller-than-wide shape, metastatic lymph nodes). Non-marked hypoechoic nodules were classified as TIRADS 3. Results: Thyroid nodules were classified as TIRADS 2, 3, 4A, 4B, and 5 in 14.5, 57.5, 14.2, 8.1, and 5.7%, respectively. Only histopathologic results (125 nodules underwent surgery) and highly specific cytology results (Bethesda II, VI) were accepted as a standard of reference, forming a sub-cohort of 562/795 nodules (70.7%). Malignancy was found in 7.7%. Overall, M-TIRADS showed sensitivity/specificity of 93.02/81.31%, and for PPV/NPV, these were 29.2/99.29%, respectively (OR—18.62). Irregular margins showed the highest sensitivity and specificity (75.68/93.74%, respectively). In TIRADS 3 category, 37.2% nodules were isoechoic, 6.6% hyperechoic, and 52.2% hypoechoic (there was no difference of malignancy risk in hypoechoic nodules between M-TIRADS and Kwak systems—0.9 vs. 0.8, respectively). Accuracy of M-TIRADS classification in this cohort was 78.26% vs. 48.11% for Kwak. Conclusions: The non-marked hypoechoic nodule pattern correlated with low risk of malignancy; classification of these nodules as TIRADS 3 significantly improved the predictive value and overall accuracy of the proposed M-TIRADS scoring with malignancy risk increase in TIRADS 4 categories by 20%; and no significant alteration of malignancy risk in TIRADS 3 could contribute to reducing overdiagnosis, obviating the need for FNA. Full article

Objective: The American College of Radiology (ACR) and the European Thyroid Association (EU) have proposed two scoring systems for thyroid nodule classification. Here, we compared the ability of the two systems in triaging thyroid nodules for fine-needle aspiration (FNA) and tested the putative role of an approach that combines ultrasound features and cytology for the detection of malignant nodules. Design and Methods: The scores obtained with the ACR and EU Thyroid Imaging Reporting and Data Systems (TIRADS) from a prospective series of 480 thyroid nodules acquired from 435 subjects were compared to assess their performances in FNA triaging on the final cytological diagnosis. The US features that showed the highest contribution in discriminating benign nodules from malignancies were combined with cytology to improve its diagnostic performance. Results: FNA was recommended on 46.5% and 51.9% of the nodules using the ACR and EU-TIRADS scores, respectively. The ACR system demonstrated a higher specificity as compared to the EU-TIRADS (59.0% vs. 52.4%, p = 0.0012) in predicting ≥ TIR3A/III (SIAPEC/Bethesda) nodules. Moreover, specific radiological features (i.e., echogenic foci and margins), combined with the cytological classes improved the specificity (97.5% vs. 91%, p < 0.0001) and positive predictive values (77.5% vs. 50.7%, p < 0.0001) compared to cytology alone, especially in the setting of indeterminate nodules (TIR3A/III and TIR3B/IV), maintaining an excellent sensitivity and negative predictive value. Conclusions: The ACR-TIRADS system showed a higher specificity compared to the EU-TIRADS in triaging thyroid nodules. The use of specific radiological features improved the diagnostic ability of cytology. Full article

Semantic features are very important for machine learning-based drug name recognition (DNR) systems. The semantic features used in most DNR systems are based on drug dictionaries manually constructed by experts. Building large-scale drug dictionaries is a time-consuming task and adding new drugs to existing drug dictionaries immediately after they are developed is also a challenge. In recent years, word embeddings that contain rich latent semantic information of words have been widely used to improve the performance of various natural language processing tasks. However, they have not been used in DNR systems. Compared to the semantic features based on drug dictionaries, the advantage of word embeddings lies in that learning them is unsupervised. In this paper, we investigate the effect of semantic features based on word embeddings on DNR and compare them with semantic features based on three drug dictionaries. We propose a conditional random fields (CRF)-based system for DNR. The skip-gram model, an unsupervised algorithm, is used to induce word embeddings on about 17.3 GigaByte (GB) unlabeled biomedical texts collected from MEDLINE (National Library of Medicine, Bethesda, MD, USA). The system is evaluated on the drug-drug interaction extraction (DDIExtraction) 2013 corpus. Experimental results show that word embeddings significantly improve the performance of the DNR system and they are competitive with semantic features based on drug dictionaries. F-score is improved by 2.92 percentage points when word embeddings are added into the baseline system. It is comparative with the improvements from semantic features based on drug dictionaries. Furthermore, word embeddings are complementary to the semantic features based on drug dictionaries. When both word embeddings and semantic features based on drug dictionaries are added, the system achieves the best performance with an F-score of 78.37%, which outperforms the best system of the DDIExtraction 2013 challenge by 6.87 percentage points. Full article