High-frequency shear wave MR elastography of parotid glands: custom driver design and preliminary results.

Parotid glands are one of the most common sites for salivary gland tumors. Conventional imaging techniques have limited usefulness in the quantitative assessment of the parotid glands, making it difficult to differentiate between healthy tissue and tumors, as well as between benign and malignant tumors. Magnetic resonance elastography (MRE) is a non-invasive technique that may potentially overcome these limitations. Nevertheless, due to the size of the parotid gland, increased elastographic resolution is required. This may be achieved by applying shear waves at higher frequencies. However, it also results in stronger attenuation, making the illumination of the parotid challenging. Here, we describe a novel passive driver tailored to the anatomy of the human face, which minimizes the distance shear waves need to travel from the source to the area of interest and thus decreases shear wave attenuation, making high-frequency shear wave MRE feasible.

A presentation of basal cell adenoma: a case report.

BACKGROUND: While statistically rare in comparison to other head and neck tumours, parotid gland swellings are often encountered in clinical practice where one of the primary goals of examination becomes distinction between benign and malignant lesions. Hallmarks of malignancy are characterized by a female preponderance, history of radiation exposure, a positive family history, and clinical features like heterogenous consistency, fixity to skin/underlying tissues and involvement of facial nerve. CASE PRESENTATION: Here we present a case of parotid swelling in a 72-year old gentleman from south India that had a curious amalgamation of both benign and malignant features. CONCLUSIONS: While benign, the risk of malignant transformation and rare multicentric occurrence indicates a need to keep basal cell adenoma in mind in case of parotid swellings and their surgical management.

Rare Presentation of Acinic Cell Carcinoma in the Parotid Gland.

Acinic cell carcinoma (ACC) arising in the salivary glands is a rare tumor. It is a low-grade malignant salivary gland tumor. It is predominantly seen in females and occurs in the fifth and sixth decades of life. It is mostly located in the parotid gland. ACC has a significant potential for recurrence and metastasis. Therefore, long-term follow-up is necessary after treatment. Here, a 28-year-old male presented with right preauricular swelling for 9 months. Ultrasound of the head-and-neck region and fine-needle aspiration cytology of preauricular swelling suggest the diagnosis of neoplasm in the parotid gland, most probably a benign tumor. After that, a total parotidectomy with facial nerve preservation was performed. On histopathological and immunohistochemical study was consistent with the diagnosis of ACC in the parotid gland.

RésuméLe carcinome à cellules aciniques (ACC) survenant dans les glandes salivaires est une tumeur rare. Il s'agit d'une tumeur de la glande salivaire maligne à faible teneur. Il est principalement observé chez les femmes et se produit dans les cinquième et sixième décennies de la vie. Il est principalement situé dans la glande parotide. ACC a un potentiel important pour la récidive et les métastases. Par conséquent, un suivi à long terme est nécessaire après le traitement. Ici, un homme de 28 ans a présenté un gonflement préauriculaire droit pendant 9 mois. Échographie de la région de la tête - et du pic et de la cytologie de l'aspiration fine Le diagnostic du néoplasme dans la glande parotide, très probablement une tumeur bénigne. Après cela, une parotidectomie totale avec préservation du nerf facial a été effectuée. Sur l'étude histopathologique et immunohistochimique, était conforme au diagnostic de l'ACC dans la glande parotide.

[Ultrasonographic features of Castleman disease in the head and neck: a report of 7 cases].

PURPOSE: To summarize the ultrasonographic features of head and neck Castleman disease (CD), and to clarify its diagnostic key points. METHODS: Seven patients with head and neck CD confirmed by histopathology were collected from Shanghai Ninth People's Hospital, Shanghai Jiao Tong University School of Medicine. The clinical features and ultrasound findings of the patients were evaluated. RESULTS: Of the 7 patients (1 male and 6 females), the mean age at diagnosis was 31.4 years (7-60 years). All the cases were hyaline vascular type. On ultrasound, 3 lesions (42.9%) were located in the parotid gland, 4 lesions(57.1%) in the neck. All the lesions presented as a solitary, well-defined and solid mass without calcification. The echogenicity was markedly hypoechoic in 1 case(14.3%) and hypoechoic in 6 cases (85.7%). Of the 7 CD cases, 4 cases (57.1%) were heterogeneous masses with linear echogenic septa. All lesions had mixed pattern in vascularity on color Doppler sonography. CONCLUSIONS: Most CDs in the head and neck represent as a markedly hypoechoic or hypoechoic lesion with mixed pattern in vascularity. The neoplasm may be characterized by the presence of linear echogenic septa within the mass.

Low Radiation Doses in Oncocytic Lesions of the Parotid Gland: A Double-edged Sword. A Comprehensive Review.

Oncocytic lesions represent a group of benign and potentially precancerous tumors characterized by the accumulation of oncocytes, which are large, granular, and eosinophilic cells. Diagnosing oncocytic lesions in the parotid gland typically involves a combination of imaging techniques, such as ultrasound, Computed Tomography (CT) scans, and Magnetic Resonance Imaging (MRI). Fine-needle aspiration (FNA) biopsy with histopathological examination remains the primary diagnostic tool for these lesions. Accurate diagnosis is crucial for appropriate management decisions. Treatment options for oncocytic lesions in the parotid gland include surgery, conservative management, and radiation therapy (RT). However, in the head and neck region, radiation doses can be a double-edged sword. While RT is a treatment modality, low radiation doses can promote the development of oncocytic lesions in the parotid gland. The prognosis for patients with oncocytic lesions is generally favorable, especially when the lesions are benign and appropriately managed. Current research focuses on the molecular mechanisms underlying oncocytic lesions in response to low-dose radiation exposure. The development of these lesions following low radiation doses represents a significant clinical concern. This manuscript provides a comprehensive overview of the current knowledge regarding oncocytic lesions in the parotid gland, including risk factors, diagnosis, treatment options, and ongoing research, offering valuable insights for clinicians and researchers.

Warthin tumor concomitant with mantle cell lymphoma: a case report and review of literature.

RATIONALE: Warthin tumor (WT) is the second most common benign tumor in salivary gland. It has a slow growth rate and most frequently occurs in the parotid gland. Most patients present with an incidental finding of a painless mass inferior/anterior to the ear. Besides the epithelial component of the tumor, WT is characteristically associated with lymphoid stroma that is considered benign. While there have been a few reports of malignant transformation of the lymphoid components in WT, cases of WT concomitant with mantle cell lymphoma (MCL) are extremely rare. To the best of our knowledge, two cases have been described in the English literature. Herein, we report a case of WT concomitant with MCL in a 70-year-old female patient, and emphasize the importance of careful examination of lymphoid stroma in WT so that concurrent lymphoma is not missed. PATIENT CONCERNS: A 70-year-old Chinese woman with a 40-year history of cigarette smoking presented with a one year history of a right submaxillary mass with recent enlargement. DIAGNOSIS: Cervical ultrasound (US) and computed tomography (CT) scans of the neck revealed a well-circumscribed mass in the right parotid with a maximum diameter of 3.1 cm. Surgical resection of the mass was performed. Histopathological examination revealed a characteristic double-layer of neoplastic epithelium with prominent lymphoid stroma, suggesting WT. In addition, morphology and immunohistochemistry studies confirmed the coexistence of MCL. Thereafter, the final diagnosis of this case was WT concomitant with MCL. INTERVENTIONS: The patient was staged as stage I after clinical assessment. Due to the slow growth of parotid lesions, close observation was decided with periodic clinical and radiological monitoring. OUTCOMES: Currently, the patient demonstrates a stable disease by clinical evaluation. LESSONS: To the best of our knowledge, reported cases of WT concomitant with MCL are very rare. This case highlights the importance of a comprehensive assessment of the lymphoid stroma of WT to avoid missed diagnosis of a lymphoma component in a collision tumor.

Minimum Fascia-Tumour Distance for Differentiating Deep Lobe From Superficial Lobe Benign Parotid Tumours: A Retrospective Study and Meta-Analysis.

BACKGROUND: This study was designed to evaluate the diagnostic efficacy of the minimum fascia-tumour distance (MFTD) in distinguishing deep-lobe benign parotid tumours from superficial-lobe tumours through both an original study and a meta-analysis. METHODS: In this study, we performed a retrospective analysis of data from 91 patients who had been diagnosed with benign parotid tumours. The MFTD values were sourced from preoperative ultrasound examinations. The locations of these tumours were confirmed through surgical findings. We assessed the diagnostic accuracy of MFTD by utilising receiver operating characteristic (ROC) curves. Additionally, we conducted a systematic review of the pertinent literature and performed a diagnostic meta-analysis to ascertain the overall diagnostic efficacy of MFTD in identifying benign parotid tumours. RESULTS: Patients with tumours in the deep lobe had a significantly greater MFTD than patients with tumours in the superficial lobe. Using a cutoff value of 3.50 mm for MFTD, we found an AUC of 0.93, a sensitivity of 81.8%, and a specificity of 98.8%. Our meta-analysis included seven studies covering a total of 1689 tumours. The pooled values for sensitivity, specificity, and diagnostic odds ratio (OR) of MFTD were 81.0%, 89.0%, and 32.2, respectively. The AUC of the summarised ROC curve of MFTD was 0.90. CONCLUSION: The MFTD demonstrated reliable diagnostic accuracy in identifying deep-lobe benign parotid tumours and may be incorporated into standard evaluations before parotidectomy.

Parotid saliva and blood biomarkers in juvenile idiopathic arthritis in relation to temporomandibular joint magnetic resonance imaging findings.

BACKGROUND: Juvenile idiopathic arthritis (JIA) often affects the temporomandibular joint (TMJ) caused by an abnormal immune system that includes overactive inflammatory processes. Salivary biomarkers may be a powerful tool that can help establishing diagnosis, prognosis and monitor disease progress. OBJECTIVE: The objective was to investigate biomarkers in parotid saliva and blood plasma in relation to temporomandibular joint (TMJ) magnetic resonance imaging (MRI) findings in patients with JIA and healthy individuals. METHODS: Forty-five children aged 6 to 16 years with JIA and 16 healthy age- and sex-matched controls were included. Unstimulated parotid saliva samples and venous blood were collected. Biochemical analyses were performed for the cytokine biomarkers. The participants underwent MR imaging of the TMJs, where changes in the inflammatory and the damage domains were assessed. RESULTS: In the JIA patients, lower concentrations of IL-6R and gp130 were found in parotid saliva than in plasma. Higher concentrations of IL-6 were found in parotid saliva than in plasma. IL-6, IL-6R and gp130 in parotid saliva explained the presence of bone marrow oedema and effusion in the JIA patients. CONCLUSIONS: This study suggests that the IL-6 family in parotid saliva is associated with TMJ bone marrow oedema and effusion in patients with JIA, suggesting that IL-6 has promising properties as a parotid saliva biomarker for TMJ inflammatory activity.

Carcinoma ex-pleomorphic adenoma of parotid gland presenting as a gigantic cervicofacial mass.

When neglected for a long time, salivary gland pleomorphic adenoma (PA) can attain a considerable size, increasing the patient's morbidity along with the risk of malignant transformation. Very few case reports are available describing PA of the parotid glands presenting as a large cervicofacial mass. We report a case of epithelial myoepithelial carcinoma -a rare subtype of carcinoma ex-PA (Ca-Ex-PA) of non-luminal differentiation, that developed over a long period in a primary PA of the parotid gland and presented as a giant cervicofacial mass.

A case of renal sarcoidosis complicated by parotid gland and uterine lesions.

BACKGROUND: Sarcoidosis is a systemic disease that can affect multiple organs. While pulmonary sarcoidosis is most commonly observed, renal sarcoidosis occurs less frequently. We herein report a case of sarcoidosis with an exceptionally rare distribution including renal lesions. CASE PRESENTATION: A 51-year-old Japanese female was referred because of bilateral parotid swelling and renal dysfunction. Computed tomography scan showed the swelling of bilateral kidneys, parotid glands, and uterus. Ga scintigraphy also showed remarkable accumulation in these organs. Renal biopsy and cytological evaluations of parotid gland and uterus were performed and she was diagnosed as sarcoidosis of these organs. Treatment was initiated with prednisolone 40 mg/day and then renal dysfunction subsequently improved. In addition, the swelling of parotid glands and uterus improved and Ga accumulation in each organ had disappeared. CONCLUSION: This is a first case of renal sarcoidosis complicated by parotid glands and uterus lesions. Pathological findings and the reactivity observed in Ga scintigraphy indicated the presence of lesions in these organs.

Evaluation of the anterior processes of the parotid gland: an ultrasonographic study.

OBJECTIVES: This study aimed to determine the prevalence of anterior extensions of the parotid gland (AEPG), namely the accessory parotid gland (APG) and the facial process of the parotid gland (FP), using ultrasonography. STUDY DESIGN: A total of 338 parotid glands were scanned bilaterally. APG was defined as a soft tissue mass with the same echogenic features as the main parotid gland (MPG) and not in contact with it, while FP was defined as an extension that exceeded the anterior border of the mandibular ramus and was continuous with the MPG. The anteroposterior, mediolateral, superoinferior dimensions and the mean distance from the MPG to the APG were measured. RESULTS: The prevalence of APG and FP were 19.5% and 36%, respectively, resulting in an AEPG prevalence of 55.6%. The presence of APG was statistically higher in females than in males (p = 0.039). The mean anteroposterior, mediolateral, and superoinferior dimensions of the APG were 18.1 ± 0.57 mm, 0.35 ± 0.17 mm, and 12.3 ± 0.36 mm, respectively, and the mean distance from the MPG was measured as 12.1 ± 0.87 mm. CONCLUSION: This study can raise awareness among clinicians about the presence of AEPG in the differential diagnosis of mid-cheek masses.

Lymphoma presenting as preauricular tumor in unilateral parotid gland agenesis: a case report and review of literature.

BACKGROUND: Parotid gland agenesis is a rare, congenital, usually asymptomatic disorder. Until now, only 24 cases with unilateral, incidentally found, parotid gland agenesis have been described. Here, we present the first reported case of an ipsilateral preauricular neoplasm in a patient with unilateral parotid gland agenesis. During surgery, the position of the greater auricular- and facial nerves was documented. Furthermore, we performed the first sialendoscopy for this rare disorder to assess the number of duct branches, which might be indicative of the abundance of parotid tissue. Moreover, we looked for sialendoscopic characteristic features that could aid in identifying these patients in the ambulatory setting. CASE PRESENTATION: A 50-year-old Greek man presented with a painless, slowly enlarging mass in the right parotid space. Magnetic resonance imaging revealed a complete absence of the right parotid gland without accessory parotid tissue. The right parotid gland was replaced by fatty tissue and the radiologist suggested a benign parotid tumor. Fine needle aspiration was indicative of a reactive lymph node. Sialendoscopy revealed only two branches within the right parotid duct. Surgical resection was performed through a conventional lateral parotidectomy. This revealed typical anatomic position of the greater auricular- and facial nerves despite the parotid tissue agenesis. Histopathology revealed a small lymphocytic lymphoma. CONCLUSIONS: Surgeons should feel confident to resect tumors of the parotid space in patients with parotid gland agenesis. Reduced branching observed during sialendoscopy might indicate parotid gland agenesis. Physicians should be even more cautious than usual with the watch and wait strategy in patients with tumors of parotid gland agenesis, since the probability of a tumor being a benign salivary gland tumor might be lower than usual.

Parotid Gland Segmentation Using Purely Transformer-Based U-Shaped Network and Multimodal MRI.

Parotid gland tumors account for approximately 2% to 10% of head and neck tumors. Segmentation of parotid glands and tumors on magnetic resonance images is essential in accurately diagnosing and selecting appropriate surgical plans. However, segmentation of parotid glands is particularly challenging due to their variable shape and low contrast with surrounding structures. Recently, deep learning has developed rapidly, and Transformer-based networks have performed well on many computer vision tasks. However, Transformer-based networks have yet to be well used in parotid gland segmentation tasks. We collected a multi-center multimodal parotid gland MRI dataset and implemented parotid gland segmentation using a purely Transformer-based U-shaped segmentation network. We used both absolute and relative positional encoding to improve parotid gland segmentation and achieved multimodal information fusion without increasing the network computation. In addition, our novel training approach reduces the clinician's labeling workload by nearly half. Our method achieved good segmentation of both parotid glands and tumors. On the test set, our model achieved a Dice-Similarity Coefficient of 86.99%, Pixel Accuracy of 99.19%, Mean Intersection over Union of 81.79%, and Hausdorff Distance of 3.87. The purely Transformer-based U-shaped segmentation network we used outperforms other convolutional neural networks. In addition, our method can effectively fuse the information from multi-center multimodal MRI dataset, thus improving the parotid gland segmentation.

Swin MoCo: Improving parotid gland MRI segmentation using contrastive learning.

BACKGROUND: Segmentation of the parotid glands and tumors by MR images is essential for treating parotid gland tumors. However, segmentation of the parotid glands is particularly challenging due to their variable shape and low contrast with surrounding structures. PURPOSE: The lack of large and well-annotated datasets limits the development of deep learning in medical images. As an unsupervised learning method, contrastive learning has seen rapid development in recent years. It can better use unlabeled images and is hopeful to improve parotid gland segmentation. METHODS: We propose Swin MoCo, a momentum contrastive learning network with Swin Transformer as its backbone. The ImageNet supervised model is used as the initial weights of Swin MoCo, thus improving the training effects on small medical image datasets. RESULTS: Swin MoCo trained with transfer learning improves parotid gland segmentation to 89.78% DSC, 85.18% mIoU, 3.60 HD, and 90.08% mAcc. On the Synapse multi-organ computed tomography (CT) dataset, using Swin MoCo as the pre-trained model of Swin-Unet yields 79.66% DSC and 12.73 HD, which outperforms the best result of Swin-Unet on the Synapse dataset. CONCLUSIONS: The above improvements require only 4 h of training on a single NVIDIA Tesla V100, which is computationally cheap. Swin MoCo provides new approaches to improve the performance of tasks on small datasets. The code is publicly available at https://github.com/Zian-Xu/Swin-MoCo.

Late-xerostomia prediction model based on 18F-FDG PET image biomarkers of the main salivary glands.

BACKGROUND AND PURPOSE: Recently, a comprehensive xerostomia prediction model was published, based on baseline xerostomia, mean dose to parotid glands (PG) and submandibular glands (SMG). Previously, PET imaging biomarkers (IBMs) of PG were shown to improve xerostomia prediction. Therefore, this study aimed to explore the potential improvement of the additional PET-IBMs from both PG and SMG to the recent comprehensive xerostomia prediction model (i.e., the reference model). MATERIALS AND METHODS: Totally, 540 head and neck cancer patients were split into training and validation cohorts. PET-IBMs from the PG and SMG, were selected using bootstrapped forward selection based on the reference model. The IBMs from both the PG and SMG with the highest selection frequency were added to the reference model, resulting in a PG-IBM model and a SMG-IBM model which were combined into a composite model. Model performance was assessed using the area under the curve (AUC). Likelihood ratio test compared the predictive performance between the reference model and models including IBMs. RESULTS: The final selected PET-IBMs were 90th percentile of the PG SUV and total energy of the SMG SUV. The additional two PET-IBMs in the composite model improved the predictive performance of the reference model significantly. The AUC of the reference model and the composite model were 0.67 and 0.69 in the training cohort, and 0.71 and 0.73 in the validation cohort, respectively. CONCLUSION: The composite model including two additional PET-IBMs from PG and SMG improved the predictive performance of the reference xerostomia model significantly, facilitating a more personalized prediction approach.

Potential predictive value of IVIM MR for xerostomia in nasopharyngeal carcinoma.

BACKGROUND AND PURPOSE: Xerostomia, caused by radiation-induced parotid damage, is the most commonly reported radiotherapy (RT) complication for nasopharyngeal carcinoma (NPC). The purpose of this study was to evaluate the value of intravoxel incoherent motion (IVIM) MR in monitoring radiation-induced parotid gland damage and predicting the risk of xerostomia. METHODS: Fifty-four NPC patients were enrolled and underwent at least three IVIM MR scans: before (pre-RT), after 5 fractions of (5th-RT), halfway through (mid-RT), and after RT (post-RT). The degree of xerostomia patients was assessed before each MR examination. Furthermore, the time when patients first reported xerostomia symptoms was recorded. The changes in IVIM parameters throughout RT, as well as the relationships between IVIM parameters and xerostomia, were analysed. RESULT: All IVIM parameters increased significantly from pre-RT to post-RT (p < 0.001). The rates of D, D* and f increase increased significantly from pre-RT to mid-RT (p < 0.001), indicating that cell necrosis mainly occurs in the first half of RT. In multivariate analysis, N3 (p = 0.014), pre-D (p = 0.007) and pre-D* (p = 0.003) were independent factors influencing xerostomia. D and f were significantly higher at 5th-RT than at pre-RT (both p < 0.05). IVIM detected parotid gland injury at 5th-RT at an average scanning time of 6.18 ± 1.07 days, earlier than the 11.94 ± 2.61 days when the patient first complained of xerostomia according to the RTOG scale (p < 0.001). CONCLUSIONS: IVIM MR can dynamically monitor radiation-induced parotid gland damage and assess it earlier and more objectively than RTOG toxicity. Moreover, IVIM can screen people at risk of more severe xerostomia early.

Multispectral optoacoustic tomography of benign parotid tumors in vivo: a prospective observational pilot study.

Parotid lumps are a heterogeneous group of mainly benign but also malignant tumors. Preoperative imaging does not allow a differentiation between tumor types. Multispectral optoacoustic tomography (MSOT) may improve the preoperative diagnostics. In this first prospective pilot trial the ability of MSOT to discriminate between the two most frequent benign parotid tumors, pleomorphic adenoma (PA) and Warthin tumor (WT) as well as to normal parotid tissue was explored. Six wavelengths (700, 730, 760, 800, 850, 900 nm) and the parameters deoxygenated (HbR), oxygenated (HbO2), total hemoglobin (HbT), and saturation of hemoglobin (sO2) were analyzed. Ten patients with PA and fourteen with WT were included (12/12 female/male; median age: 51 years). For PA, the mean values for all measured wave lengths as well as for the hemoglobin parameters were different for the tumors compared to the healthy parotid (all p < 0.05). The mean MSOT parameters were all significantly higher (all p < 0.05) in the WT compared to healthy parotid gland except for HbT and sO2. Comparing both tumors directly, the mean values of MSOT parameters were not different between PA and WT (all p > 0.05). Differences were seen for the maximal MSOT parameters. The maximal tumor values for 900 nm, HbR, HbT, and sO2 were lower in PA than in WT (all p < 0.05). This preliminary MSOT parotid tumor imaging study showed clear differences for PA or WT compared to healthy parotid tissue. Some MSOT characteristics of PA and WT were different but needed to be explored in larger studies.

An ultrasound-based ensemble machine learning model for the preoperative classification of pleomorphic adenoma and Warthin tumor in the parotid gland.

OBJECTIVES: The preoperative classification of pleomorphic adenomas (PMA) and Warthin tumors (WT) in the parotid gland plays an essential role in determining therapeutic strategies. This study aims to develop and validate an ultrasound-based ensemble machine learning (USEML) model, employing nonradiative and noninvasive features to differentiate PMA from WT. METHODS: A total of 203 patients with histologically confirmed PMA or WT who underwent parotidectomy from two centers were enrolled. Clinical factors, ultrasound (US) features, and radiomic features were extracted to develop three types of machine learning model: clinical models, US models, and USEML models. The diagnostic performance of the USEML model, as well as that of physicians based on experience, was evaluated and validated using receiver operating characteristic (ROC) curves in internal and external validation cohorts. DeLong's test was used for comparisons of AUCs. SHAP values were also utilized to explain the classification model. RESULTS: The USEML model achieved the highest AUC of 0.891 (95% CI, 0.774-0.961), surpassing the AUCs of both the US (0.847; 95% CI, 0.720-0.932) and clinical (0.814; 95% CI, 0.682-0.908) models. The USEML model also outperformed physicians in both internal and external validation datasets (both p < 0.05). The sensitivity, specificity, negative predictive value, and positive predictive value of the USEML model and physician experience were 89.3%/75.0%, 87.5%/54.2%, 87.5%/65.6%, and 89.3%/65.0%, respectively. CONCLUSIONS: The USEML model, incorporating clinical factors, ultrasound factors, and radiomic features, demonstrated efficient performance in distinguishing PMA from WT in the parotid gland. CLINICAL RELEVANCE STATEMENT: This study developed a machine learning model for preoperative diagnosis of pleomorphic adenoma and Warthin tumor in the parotid gland based on clinical, ultrasound, and radiomic features. Furthermore, it outperformed physicians in an external validation dataset, indicating its potential for clinical application. KEY POINTS: • Differentiating pleomorphic adenoma (PMA) and Warthin tumor (WT) affects management decisions and is currently done by invasive biopsy. • Integration of US-radiomic, clinical, and ultrasound findings in a machine learning model results in improved diagnostic accuracy. • The ultrasound-based ensemble machine learning (USEML) model consistently outperforms physicians, suggesting its potential applicability in clinical settings.