Ultrasound imaging is widely used across cardiology, hepatology, obstetrics, breast and thyroid imaging, and emergency care because it is real-time, noninvasive, and relatively accessible. However, its diagnostic performance remains influenced by operator experience, image quality, scanner settings, and interpretation variability. Artificial intelligence (AI) has emerged as a promising support tool for ultrasound, assisting with image acquisition, quality assessment, view classification, segmentation, measurement, lesion characterization, and structured reporting. This narrative review summarizes recent developments in AI-assisted ultrasound imaging, emphasizing its technical foundations, clinical applications, validation challenges, and relevance to Gulf health care systems. Current evidence suggests that AI may improve workflow efficiency, reduce interobserver variability, and support diagnostic decision-making in selected ultrasound tasks, particularly when models are trained and tested on large, diverse data sets. Nevertheless, the clinical readiness of many AI tools remains constrained by retrospective study designs, single-center data sets, limited external validation, vendor-dependent image variability, and insufficient prospective evaluation. These limitations are particularly salient in Gulf health care, where ultrasound services are delivered across heterogeneous public, private, military, and academic institutions that use different equipment, workflows, and operator training backgrounds. Gulf countries are well-positioned to adopt AI-enabled ultrasound because of the ongoing digital health transformation and emerging regulatory frameworks, including the Saudi SFDA guidance and the UAE AI governance initiatives. However, responsible implementation will require region-specific, multicenter, multivendor validation, transparent reporting of model performance and failure cases, clinician training, and privacy-preserving data governance. AI should therefore be viewed not as a replacement for ultrasound professionals but as a decision-support technology that may improve consistency, efficiency, and diagnostic confidence when carefully validated in real-world clinical settings.
Thyroid nodules are frequently encountered during carotid Doppler ultrasonography, yet their clinical significance remains uncertain. This prospective study evaluated the prevalence, management, and clinical yield of thyroid nodules detected during opportunistic thyroid assessment performed alongside carotid Doppler examinations. A total of 510 consecutive adult patients were enrolled between November 2025 and May 2026. Thyroid nodules were classified according to the American College of Radiology Thyroid Imaging Reporting and Data System (ACR TI-RADS). Follow-up recommendations, fine-needle aspiration biopsy (FNAB) rates, cytopathological outcomes, and biopsy adherence were analyzed. Thyroid nodules were detected in 338 patients (66.3%). Follow-up ultrasonography was recommended in 58 patients (17.2%), and FNAB was recommended in 37 (11.0%). Biopsy was performed in 21 patients (56.8%). Three malignancies were identified, yielding an overall malignancy rate of 0.6% (3/510). Patients who did not undergo recommended biopsy were significantly older than those who underwent biopsy (P=0.032), and biopsy adherence was lower among rural residents (P=0.039). Although thyroid nodules were highly prevalent, the overall malignancy yield was low. An ACR TI-RADS-guided selective reporting strategy may reduce unnecessary investigations while maintaining detection of clinically significant thyroid cancer.
Isolated fallopian tube torsion (IFTT) is a rare form of adnexal torsion. A 2024 review described fewer than 150 cases over several decades. The clinical presentation of IFTT is highly variable and nonspecific, which may result in delayed diagnosis. Most patients present with acute unilateral abdominal pain with nausea and/or vomiting, while other patients may present with more gradual symptoms. Risk factors for IFTT include adnexal lesions, hydrosalpinx, previous pelvic surgery, and pelvic inflammatory disease. Ultrasound is particularly useful in identifying the imaging signs of IFTT, which include normal ovarian size and flow, a distended cylindrical structure that tapers to a "beak" adjacent to the ovary, thickened endosalpingeal folds resulting in a "cogwheel" appearance, and, lastly, a twisted vascular pedicle. We present an imaging-focused case in which the diagnosis of IFTT using ultrasound, following equivocal CT, resulted in prompt and confirmatory surgical management. Our goal is to raise awareness of IFTT as a possibility in females with abdominal pain and to discuss ultrasound characteristics that may increase clinical suspicion and diagnostic confidence. Prompt recognition and surgical management may prevent tubal necrosis and loss of fertility.
Adult acquired buried penis (AABP) disease is a risk factor for penile cancer. For a subset of patients, the severity of their disease precludes complete physical examination of the penile shaft due to the inability to exhume the penis. Thus, providers are unable to identify penile lesions, which may alter surgical plans to include a more extensive oncologic treatment including penectomy. We aimed to assess the use of preoperative penile ultrasound in patients with adult acquired buried penis before surgical intervention. We performed a retrospective cohort review of ten patients whose phalli were unable to be fully exposed for examination and underwent penile ultrasonography before AABP repair with a single reconstructive urologist. Radiographic images were performed and reviewed by a single radiologist with experience interpreting genitourinary imaging. Surgical pathology was reviewed by a genitourinary pathologist. Of the ten patients who underwent preoperative ultrasound, two had physical exam findings concerning for possible penile masses. Ultrasonographic evaluation of these patients revealed tissue heterogeneity and color doppler hyperemia of the penis. Surgical specimen pathology review of both patients demonstrated invasive penile cancer. The remaining eight patients without concerning examination findings had unremarkable ultrasound findings and subsequent benign surgical pathology. Preoperative penile ultrasound may be a valuable adjunct in the evaluation of penile lesions in patients with AABP whose anatomy does not allow for complete physical examination and clinical staging.
This study aims to establish gestational age and sex-specific nomograms for fetal corpus callosum (CC) length measured by two-dimensional (2D) and three-dimensional (3D) ultrasonography, and to evaluate measurement differences between these imaging techniques during the second trimester. The retrospective single-center study included singleton pregnancies that underwent detailed second-trimester ultrasonography between 19 and 25 gestational weeks. Fetal CC length was measured on the midsagittal plane using both 2D and 3D ultrasonography. Measurements were performed independently by 2 experienced perinatologists. Gestational age-specific percentile values (5th, 25th, 50th, 75th, and 95th) were generated separately for each modality. Agreement between observers and between imaging techniques was assessed using intraclass correlation coefficients. A total of 211 fetuses were included. Fetal CC length showed a progressive increase with advancing gestational age in both imaging modalities. No significant differences were observed between male and female fetuses. Measurements obtained by 3D ultrasonography were consistently higher than those obtained by 2D ultrasonography, with the difference reaching statistical significance. & Idot;n conclusion, the present study provides modality-specific nomograms for fetal CC length during mid-gestation and demonstrates that 3D ultrasonography yields higher CC measurements compared with 2D imaging. These findings support the use of separate reference ranges for each modality and suggest that 3D ultrasonography may offer practical advantages in the assessment of fetal midline brain structures.
This study assessed the diagnostic efficiency of strain elastography in differentiating benign and malignant non-mass breast lesions (NMLs) detected on ultrasonography. A total of 149 patients (mean age, 48.8±10.6 y) with histopathologically confirmed NMLs were retrospectively analyzed. Each lesion was evaluated using B-mode ultrasonography and strain elastography, and both elasticity score and strain ratio were recorded. The mean strain ratio was significantly higher in malignant lesions than in benign ones (3.2±0.96 vs. 1.8±0.9, P <0.001). Receiver operating characteristic curve analysis showed a high diagnostic accuracy for strain ratio (area under the curve=0.863; 95% CI: 0.794-0.931), with a cutoff value of 2.45 yielding 82.8% sensitivity and 77.5% specificity. Lesions with an elasticity score ≥4 were more likely to be malignant ( P <0.001). In multivariable modeling, adding elastography parameters to conventional ultrasonography improved diagnostic performance, and decision curve analysis demonstrated superior net clinical benefit for elastography-based models compared with ultrasonography alone. These findings suggest that integrating strain elastography into routine ultrasonographic assessment can enhance diagnostic accuracy and support clinical decision-making in patients with NMLs.
Endometriosis is a chronic inflammatory condition affecting 10% to 15% of reproductive-aged women. The urinary tract is the second most common extragenital site of endometriosis after the gastrointestinal tract, with a prevalence of 15% to 50% of women with deep endometriosis (DE). The urinary bladder is the most common site of urinary tract involvement (85%), followed by the ureter (10%), kidney (4%), and urethra (2%). Urinary bladder (anterior compartment) and ureter (mediolateral compartment) involvement are considered different disease entities. Patients with bladder involvement are more symptomatic with dysuria, urinary frequency, and recurrent urinary tract infections. Ureteral involvement is more commonly due to extrinsic compression, but may be intrinsic, involving the ureteral mucosa or muscularis. Hematuria is a rare presenting symptom of both bladder and ureteral involvement. Malignant transformation of urinary tract endometriosis is rare; however, DE involvement of the urinary tract may be mistaken for malignancy. Radiologists need a high index of suspicion for endometriosis in reproductive-aged women, and recognition of urinary tract involvement is important for timely treatment.
Given the realities of current radiologist and sonographer staffing shortages, increasing caseloads, productivity metrics for performance, and pressure for faster turnaround times in ultrasound, this survey polled members of the Society of Radiologists in Ultrasound. The membership includes academic and private radiologist practitioners, experts, and directors in the field. The goal was to define the state of ultrasound practice today in various settings, understand the tradeoffs between speed and accuracy, and offer solutions that can mitigate these trends. An anonymous survey regarding specific details of individual ultrasound practice was circulated to the 500 members of the Society of Radiologists in Ultrasound, with 101 responses tallied in total. The majority of respondents reported a difference in their practice of ultrasound, including increased case volumes, decreased review of cases by radiologists before patients leave the department, and decreased time for academics and teaching. Results of this survey reveal threats to academic practice, oversight, and quality of scan reports, resident education, and staff retention. Possible solutions to mitigate these trends are discussed.
The 2025 FIGO Cancer Report updates recommendations for the diagnosis, staging, and management of the 6 major gynecologic malignancies, including cervical, endometrial, vaginal, vulvar, ovarian, and fallopian tube cancer, and gestational trophoblastic disease (GTD). The role of ultrasound differs considerably across the 6 cancers. In cases of endometrial and GTD, ultrasound is used to guide initial diagnosis, risk stratification, and surgical planning. In postmenopausal bleeding, a transvaginal ultrasound can assess endometrial thickness and determine the need for further evaluation. In cervical cancer, transrectal ultrasound allows local tumor assessment and brachytherapy guidance. Ovarian and fallopian tube cancers rely on ultrasound as the first-line tool for adnexal mass assessment and contribute to the Risk of Malignancy Index. For vaginal cancer, sonography has a complementary role in the anatomic assessment of the primary tumor. In vulvar cancer, ultrasound use is limited, as it is mainly for lymph node evaluation in early disease, when used in conjunction with clinical examination and cross-sectional imaging. Although ultrasound has limitations when assessing nodal spread or distant metastases, it can be used in combination with cross-sectional imaging and serial biomarkers to provide a complete evaluation. This review aims to summarize the ultrasound-relevant content from the 2025 FIGO update reports and serve as a practical reference for radiologists.
Cardiovascular ultrasound is an operator-dependent modality that requires careful sequencing of the theoretical and practical training to optimize skills acquisition. This study compares lecture-first (LF) versus practical-first (PF) teaching strategies in undergraduate radiologic sciences students to evaluate the impact of 2 different teaching strategies on students' scores. In a randomized controlled trial, 54 students were allocated to LF (n=25) or PF (n=29) groups. The LF group received didactic lectures followed by hands-on practice, while the PF group reversed this sequence. Outcomes were measured using a multiple-choice question (MCQ) exam for theoretical knowledge and an objective structured clinical examination (OSCE) for practical skills. Evaluation was performed by blinded faculty using standardized rubrics. An exploratory analysis of the impact of supplemental practice sessions was also performed. LF students demonstrated superior overall performance (median score: 0.9 vs. 0.82, P=0.009, effect size: d=0.72) and practical skills (0.95 vs. 0.82, P=0.002, effect size: d=0.85). Theoretical knowledge scores were comparable between groups, with median scores of 0.86 for LF and 0.85 for PF (P=0.9). Students attending extra practice sessions achieved significantly higher scores across all domains (overall: 0.9 vs. 0.83, P=0.004; skills: 0.93 vs. 0.87, P=0.01). Teaching strategy in which a lecture is delivered first, followed by a practical session, was associated with improved scores compared with the design starting with a practical session. Attending additional hands-on sessions can significantly enhance students' knowledge and skills for performing ultrasound examinations. This study suggests that sequencing can influence student academic achievement and indicates that arranging practical sessions after lectures and providing extra practical sessions can enhance learning in the cardiovascular ultrasound curriculum.
This study evaluates whether contrast-enhanced ultrasound (CEUS) serves as an effective next-step imaging modality for patients with end-stage renal disease (ESRD) who have indeterminate renal lesions on computed tomography (CT), assessing its diagnostic performance and potential impact on surgical decision-making. A retrospective chart review was conducted on patients with ESRD who underwent both CEUS and CT between 2021 and 2024 within the VCU Health System. The primary lesion analyzed in each case was the one most concerning for malignancy per the radiology report. If no malignancy was suspected, the largest lesion was selected. Imaging findings from CEUS and CT were assessed based on radiology reports and compared against histopathological results and clinical follow-up. A total of 76 patients met the inclusion criteria. There was no significant difference in lesion characterization accuracy between CEUS and CT (P=0.51), lesion diameter (P=0.70), or location (P=0.87). Among the 12 patients who underwent surgery due to suspected lesion malignancy, 11 (91.7%) were referred for surgery based on CEUS findings, with 100% confirmed malignancy on histopathology. CEUS demonstrates diagnostic performance comparable to CT for characterizing indeterminate renal lesions in patients with ESRD. When used as the next step after noncontrast CT, CEUS can reliably guide surgical decision-making while minimizing contrast and radiation exposure. Early integration of CEUS into structured clinical pathways may streamline evaluation, reduce delays to definitive management, and support timely progression toward transplant eligibility. Its safety, accessibility, and cost-effectiveness make CEUS a valuable tool for guiding care in this complex population.
This study aims to establish gestational age and sex-specific nomograms for fetal corpus callosum (CC) length measured by two-dimensional (2D) and three-dimensional (3D) ultrasonography, and to evaluate measurement differences between these imaging techniques during the second trimester. The retrospective single-center study included singleton pregnancies that underwent detailed second-trimester ultrasonography between 19 and 25 gestational weeks. Fetal CC length was measured on the midsagittal plane using both 2D and 3D ultrasonography. Measurements were performed independently by 2 experienced perinatologists. Gestational age-specific percentile values (5th, 25th, 50th, 75th, and 95th) were generated separately for each modality. Agreement between observers and between imaging techniques was assessed using intraclass correlation coefficients. A total of 211 fetuses were included. Fetal CC length showed a progressive increase with advancing gestational age in both imaging modalities. No significant differences were observed between male and female fetuses. Measurements obtained by 3D ultrasonography were consistently higher than those obtained by 2D ultrasonography, with the difference reaching statistical significance. İn conclusion, the present study provides modality-specific nomograms for fetal CC length during mid-gestation and demonstrates that 3D ultrasonography yields higher CC measurements compared with 2D imaging. These findings support the use of separate reference ranges for each modality and suggest that 3D ultrasonography may offer practical advantages in the assessment of fetal midline brain structures.
Cardiovascular ultrasound is an operator-dependent modality that requires careful sequencing of the theoretical and practical training to optimize skills acquisition. This study compares lecture-first (LF) versus practical-first (PF) teaching strategies in undergraduate radiologic sciences students to evaluate the impact of 2 different teaching strategies on students' scores. In a randomized controlled trial, 54 students were allocated to LF (n=25) or PF (n=29) groups. The LF group received didactic lectures followed by hands-on practice, while the PF group reversed this sequence. Outcomes were measured using a multiple-choice question (MCQ) exam for theoretical knowledge and an objective structured clinical examination (OSCE) for practical skills. Evaluation was performed by blinded faculty using standardized rubrics. An exploratory analysis of the impact of supplemental practice sessions was also performed. LF students demonstrated superior overall performance (median score: 0.9 vs. 0.82, P=0.009, effect size: d=0.72) and practical skills (0.95 vs. 0.82, P=0.002, effect size: d=0.85). Theoretical knowledge scores were comparable between groups, with median scores of 0.86 for LF and 0.85 for PF (P=0.9). Students attending extra practice sessions achieved significantly higher scores across all domains (overall: 0.9 vs. 0.83, P=0.004; skills: 0.93 vs. 0.87, P=0.01). Teaching strategy in which a lecture is delivered first, followed by a practical session, was associated with improved scores compared with the design starting with a practical session. Attending additional hands-on sessions can significantly enhance students' knowledge and skills for performing ultrasound examinations. This study suggests that sequencing can influence student academic achievement and indicates that arranging practical sessions after lectures and providing extra practical sessions can enhance learning in the cardiovascular ultrasound curriculum.
Endometriosis is a chronic inflammatory condition affecting 10% to 15% of reproductive-aged women. The urinary tract is the second most common extragenital site of endometriosis after the gastrointestinal tract, with a prevalence of 15% to 50% of women with deep endometriosis (DE). The urinary bladder is the most common site of urinary tract involvement (85%), followed by the ureter (10%), kidney (4%), and urethra (2%). Urinary bladder (anterior compartment) and ureter (mediolateral compartment) involvement are considered different disease entities. Patients with bladder involvement are more symptomatic with dysuria, urinary frequency, and recurrent urinary tract infections. Ureteral involvement is more commonly due to extrinsic compression, but may be intrinsic, involving the ureteral mucosa or muscularis. Hematuria is a rare presenting symptom of both bladder and ureteral involvement. Malignant transformation of urinary tract endometriosis is rare; however, DE involvement of the urinary tract may be mistaken for malignancy. Radiologists need a high index of suspicion for endometriosis in reproductive-aged women, and recognition of urinary tract involvement is important for timely treatment.
Splenomegaly is an important clinical sign associated with diverse pathologies. Traditional bedside physical examination maneuvers used to detect splenomegaly demonstrate poor diagnostic performance. Point-of-care ultrasound (POCUS) is a potential alternative. We sought to compare the diagnostic performance of structured physical examination maneuvers and POCUS in the bedside detection of splenomegaly. Internal medicine and hematology inpatients at Kingston Health Sciences Centre from February 2023 to June 2024 were prospectively enrolled. Participants underwent a structured physical exam comprised of Castell's Method, Traube's Method, Nixon's Method, and one-handed palpation followed by a POCUS examination. Radiographic confirmation of splenomegaly (spleen length >12 cm) was obtained via abdominal ultrasound, computed tomography, or magnetic resonance imaging within 1 month of enrollment. The performance of each diagnostic test was determined. Of 130 patients enrolled, 55 (38.5%) had radiographically confirmed splenomegaly. POCUS demonstrated higher diagnostic performance (sensitivity 70.9%, specificity 72.0%, Matthew's correlation coefficient 0.43) compared with physical exam maneuvers. The most specific physical exam maneuver was Nixon's method (90.8%), while percussion of Traube's space had the highest sensitivity (52.7%). POCUS measurements strongly correlated with formal imaging (Spearman rho=0.630; P<0.001). The diagnostic accuracy of POCUS was reduced in patients with elevated body mass index (P=0.027) and in those scanned within 2 hours of eating (P=0.047). The POCUS examination took almost twice as long as the combined physical exam maneuvers (median 142 vs. 82). POCUS outperforms traditional physical examination in detecting splenomegaly and correlates well with formal imaging, supporting its integration into bedside practice and medical education.
The objective and purpose of this research are to establish the definition of accuracy of diagnosis for both techniques, EUS-guided fine-needle aspirations (FNA) and ultrasound-guided core biopsies (US-CNB), in the assessment of the pancreatic lesion depending on its size, position, and type of lesions. This study retrospectively reviewed the clinical records of patients who underwent US-CNB or EUS-FNA biopsy for pancreatic lesions between January 2019 and 2022. For each patient, demographic information and information on lesion size, location, and type were obtained. A total of 229 subjects were recruited into the study, with 140 in the US-CNB group and 89 subjects in the EUS-FNA group. The results revealed higher diagnostic efficiency in the US-CNB group compared with the EUS-FNA group, with efficiencies of 96.4% and 70.8%, respectively ( P <0.001). Multivariate logistic regression showed that, after adjusting for confounders, US-CNB remained an independent predictor (OR=3.051, P =0.011). After propensity score matching, US-CNB continued to demonstrate significantly higher diagnostic efficiency compared with EUS-FNA (98.4% vs. 71.0%, P <0.001). Diagnostic performance parameters were evaluated in 105 patients with histopathologically established final diagnosis and at least 1 year of follow-up. In addition, US-CNB demonstrated higher sensitivity and specificity when compared with EUS-FNA, which were 92.7% and 83%, respectively. In this retrospective cohort, US-CNB demonstrated higher sample adequacy and sensitivity compared with EUS-FNA for the diagnosis of pancreatic lesions. This advantage was particularly evident in lesions smaller than 4 cm and those located in the head-uncinate process and body of the pancreas.
This study assessed the diagnostic efficiency of strain elastography in differentiating benign and malignant non-mass breast lesions (NMLs) detected on ultrasonography. A total of 149 patients (mean age, 48.8±10.6 y) with histopathologically confirmed NMLs were retrospectively analyzed. Each lesion was evaluated using B-mode ultrasonography and strain elastography, and both elasticity score and strain ratio were recorded. The mean strain ratio was significantly higher in malignant lesions than in benign ones (3.2±0.96 vs. 1.8±0.9, P<0.001). Receiver operating characteristic curve analysis showed a high diagnostic accuracy for strain ratio (area under the curve=0.863; 95% CI: 0.794-0.931), with a cutoff value of 2.45 yielding 82.8% sensitivity and 77.5% specificity. Lesions with an elasticity score ≥4 were more likely to be malignant (P<0.001). In multivariable modeling, adding elastography parameters to conventional ultrasonography improved diagnostic performance, and decision curve analysis demonstrated superior net clinical benefit for elastography-based models compared with ultrasonography alone. These findings suggest that integrating strain elastography into routine ultrasonographic assessment can enhance diagnostic accuracy and support clinical decision-making in patients with NMLs.
Given the realities of current radiologist and sonographer staffing shortages, increasing caseloads, productivity metrics for performance, and pressure for faster turnaround times in ultrasound, this survey polled members of the Society of Radiologists in Ultrasound. The membership includes academic and private radiologist practitioners, experts, and directors in the field. The goal was to define the state of ultrasound practice today in various settings, understand the tradeoffs between speed and accuracy, and offer solutions that can mitigate these trends. An anonymous survey regarding specific details of individual ultrasound practice was circulated to the 500 members of the Society of Radiologists in Ultrasound, with 101 responses tallied in total. The majority of respondents reported a difference in their practice of ultrasound, including increased case volumes, decreased review of cases by radiologists before patients leave the department, and decreased time for academics and teaching. Results of this survey reveal threats to academic practice, oversight, and quality of scan reports, resident education, and staff retention. Possible solutions to mitigate these trends are discussed.
The 2025 FIGO Cancer Report updates recommendations for the diagnosis, staging, and management of the 6 major gynecologic malignancies, including cervical, endometrial, vaginal, vulvar, ovarian, and fallopian tube cancer, and gestational trophoblastic disease (GTD). The role of ultrasound differs considerably across the 6 cancers. In cases of endometrial and GTD, ultrasound is used to guide initial diagnosis, risk stratification, and surgical planning. In postmenopausal bleeding, a transvaginal ultrasound can assess endometrial thickness and determine the need for further evaluation. In cervical cancer, transrectal ultrasound allows local tumor assessment and brachytherapy guidance. Ovarian and fallopian tube cancers rely on ultrasound as the first-line tool for adnexal mass assessment and contribute to the Risk of Malignancy Index. For vaginal cancer, sonography has a complementary role in the anatomic assessment of the primary tumor. In vulvar cancer, ultrasound use is limited, as it is mainly for lymph node evaluation in early disease, when used in conjunction with clinical examination and cross-sectional imaging. Although ultrasound has limitations when assessing nodal spread or distant metastases, it can be used in combination with cross-sectional imaging and serial biomarkers to provide a complete evaluation. This review aims to summarize the ultrasound-relevant content from the 2025 FIGO update reports and serve as a practical reference for radiologists.
