Ultrasound Reports Research Articles

Inter-Reader Agreement for Contrast-Enhanced Ultrasound Liver Imaging Reporting and Data System Major Features and Final Categorization: A Subanalysis From a Prospective Multicenter Study.

Contrast-enhanced ultrasound (CEUS) Liver Imaging Reporting and Data System (LI-RADS) is used to definitively diagnose hepatocellular carcinoma (HCC) in patients at risk. However, the user variability associated with CEUS LI-RADS has not been validated in North American and European patients. This study aims to evaluate the inter-reader agreements of CEUS LI-RADS features for, and final categorization of, HCC in patients at risk. This retrospective multicenter clinical study used the database of a previous prospective multinational study, evaluating the accuracy of CEUS LI-RADS for HCC diagnosis in patients at risk. All cases were first evaluated by a site physician performing/supervising the CEUS examination. Randomly selected cases were re-evaluated by a blinded central reader. Final diagnosis was confirmed with the reference standard, which was a composite of imaging tests and histology. Cohen's kappa test was used to evaluate inter-reader agreement. This study included 150 liver nodules and 58.0% (87/150) were confirmed as HCC, 4.7% (7/150) non-HCC malignancies, 22.7% (34/150) had no confirmed final diagnosis, and 14.7% (22/150) were nonmalignant. Inter-reader agreements were substantial for CEUS LI-RADS categorization (κ = 0.61; 95% confidence interval [CI]: 0.51-0.71) and major features assessment (ranged κ = 0.64-0.78), LR-5 (κ = 0.65; 95% CI: 0.52-0.77), and LR-M (κ = 0.67; 95% CI: 0.44-0.90), while for LR-1 and LR-2 categorization was almost perfect (κ = 0.85; 95% CI: 0.65-1.00). Our study reported a substantial inter-reader agreement for overall CEUS LI-RADS categorization, especially for LR-5 and LR-M, and major imaging features of HCC, further confirming CEUS LI-RADS as a valuable and reliable tool for diagnosis of HCC.

Pre-referral ultrasound for cryptorchidism: Still common, still not necessary.

Evidence-based guidelines do not recommend imaging in cryptorchidism, but anecdotally most referrals include an ultrasound report. We aimed to assess the frequency, utility and burden of imaging in children referred with presumptive disorders of testicular descent, and to assess trends over a 7-year period before and after local and international guidelines have been introduced. This was a prospective cohort study of children referred to the Queensland Children'sHospital for anomalies of testicular descent between 2015-2017 and 2023-2024. Data were collected regarding demographics, referral details, imaging performed and surgical diagnosis. A total of 268 children were recruited. Ultrasound frequency has not significantly changed over time (72.8%, 2015-2017; 63.6% 2023-2024; P = 0.11). Currently, 17.6% of families are charged, and 31.9% need to take time off work, for the ultrasound. This is a significant increase from the 2015-2017 cohort. Parents report concern and anxiety, and find it traumatic for their child. Following review, the majority (65.7%) had physiologically normal testes, while 4.9% and 22.4% were diagnosed with bilateral or unilateral cryptorchidism, respectively, and 7.0% had other diagnoses. Ultrasound was concordant with the surgical diagnosis in 25.0% (2015-2017) and 30.7% (2023-2024). Ultrasound did not contribute to diagnosis nor management in any patient. Despite international and local guidelines, cryptorchid children continue to undergo ultrasound prior to referral. Such ultrasounds do not provide utility, or value for cost. We recommend avoiding their routine use for suspicion of testicular maldescent and working together to improve strategies for translating evidence-based guidelines into clinical practice.

Cesarean Scar niche: Comparison of Transvaginal Ultrasound Findings and Uterine Tissue Samples after Hysterectomy

BackgroundThe cesarean scar (CS) niche, a consequence of incomplete wound healing, can lead to gynecological complications and affect future pregnancies. While internal niches have been well-studied, external and bidirectional niches are also important. ObjectivesWe aimed to conduct a comparative analysis of the prevalence and measurements of different types of niches. Study DesignThis cross-sectional study examined 200 patients with prior cesarean delivery (CD) undergoing hysterectomy due to abnormal uterine bleeding. It evaluated internal, external, and bidirectional niche prevalence, measurements (height, residual myometrial thickness, and adjacent myometrial thickness), and classification in hysterectomy specimens compared to pre-hysterectomy TVS reports. The secondary outcome was to determine niche presence in uterine tissue samples based on the number of previous CDs. ResultsThe mean age of the study participants was 48.19 years, with 30% having undergone only one CD. Notably, 83.5% had niches detected via sonography, while 91% had niches identified in uterine tissue samples post-hysterectomy (p-value =0.008). Internal niches were common, with significant associations between niche presence and the number of previous CDs. 23.3% of patients with one CD had internal niches. Additionally, external niches were detected in 4% of hysterectomy samples but not observed in ultrasound reports. Key parameters such as RMT and AMT differed notably between ultrasound and tissue samples, highlighting discrepancies in niche detection methods. RMT ranged from 2.8-24 mm (7.68 ± 3.09 mm) in ultrasound reports versus 0-25 mm (4.28 ± 2.71 mm) in tissue samples (p < 0.001). AMT ranged from 6-29 mm (17.08 ± 4.53 mm) in ultrasound reports versus 7.5-30 mm (16 ± 5.03 mm) in tissue samples (p < 0.001). This study underscores the importance of accurate niche assessment in patients with prior CD. ConclusionsWe performed a comparative analysis of niche prevalence and measurements in gross specimens and TVS reports. The results highlight the importance of considering external and bidirectional niches, in addition to internal niches. These niches can reduce residual myometrial thickness and increase future pregnancy complications. Furthermore, we demonstrated that niche formation can occur after a single CD.

Ultrasound strain elastography to improve diagnostic performance of breast lesions by reclassifying BI-RADS 3 and 4a lesions: a multicenter diagnostic study.

To investigate the added value of strain elastography (SE) by recategorizing ultrasound (US) Breast Imaging Reporting and Data System (BI-RADS) 3 and 4a lesions. A total of 4371 patients underwent US and SE with BI-RADS 2-5 categories solid breast lesions were included from thirty-two hospitals. We evaluated the elastographic images according to elasticity scores (ES) and strain ratios (SR). Three combined methods (BI-RADS+ES, BI-RADS+SR, BI-RADS+ES+SR) and two reclassified methods were used (method one: upgrading BI-RADS 3 and downgrading BI-RADS 4a, method two: downgrading BI-RADS 4a alone). The diagnostic performance and the potential reduction of unnecessary biopsies were evaluated. Combining BI-RADS with SE had a higher area under the curve (AUC) than BI-RADS alone (0.822-0.898 vs. 0.794, P<0.01). For reclassified method one, the sensitivity, specificity, and accuracy were 99.36%, 66.70%, 78.36% for BI-RADS+ES and 98.01%, 66.45%, 77.72% for BI-RADS+SR, and 99.42%, 66.70%, 78.38% for BI-RADS+ES+SR, respectively. For reclassified method two, the sensitivity, specificity, and accuracy were 99.17%, 70.72%, 80.87% for BI-RADS+ES and 97.76%, 81.75%, 87.46% for BI-RADS+SR, and 99.23%, 69.83%, 80.32% for BI-RADS+ES+SR, respectively. Downgrading BI-RADS 4a alone had higher AUC, specificity, and accuracy (P<0.01) and similar sensitivity (P>0.05) to upgrading BI-RADS 3 and downgrading BI-RADS 4a. Combining SE with BI-RADS could help reduce unnecessary biopsies by 17.64%-55.20%. Combining BI-RADS with SE improved the diagnostic performance in distinguishing benign from malignant lesions and could decrease false-positive breast biopsy rates. Downgrading BI-RADS 4a lesions alone might be sufficient for achieving good diagnostic performance. Downgrading BI-RADS category 4a lesions alone had higher AUC, specificity, and accuracy, and similar sensitivity to upgrading or downgrading BI-RADS category 3 and 4a lesions.

Use of Natural Language Processing and Deep Learning to Analyze Kidney Ultrasound Reports and Their Correlation with CKD Diagnosis

Use of Natural Language Processing and Deep Learning to Analyze Kidney Ultrasound Reports and Their Correlation with CKD Diagnosis

Influence of Prior Imaging Review on Recommendations for Additional Diagnostic Testing: Retrospective Analysis of Imaging Reports in a Chiropractic Radiology Practice

ObjectiveThe purpose of this study was to assess the influence of prior imaging review on recommendations for additional diagnostic testing in an academic chiropractic radiology practice. A secondary aim was to explore the influence of prior imaging review on radiographic interpretation. MethodsWe retrospectively reviewed radiology reports generated from July 18, 2022, to July 18, 2023, from the Palmer College of Chiropractic main campus (Davenport, Iowa) clinic system. Imaging interpretation included an automated search for prior images in an internal picture archival and communication system (PACS). Images from regional health system databases were available and sought by radiologists when (1) unclear radiologic findings had potential clinical implications or (2) prior imaging could clarify potential problems detected in a clinical history. Data were abstracted to a secure adaptive electronic questionnaire and analyzed descriptively. ResultsWe reviewed 1712 radiographic and 165 musculoskeletal diagnostic ultrasound reports for 1552 unique individuals (811 [52.3%] females and 741 [47.7%] males) with a mean age of 42.1 years (range, 2-93 years). Prior imaging was described in 417 (22.2%) reports; 246 (58.9%) indicated images from internal PACS, 192 (46.0%) indicated images from external PACS, and 21 noted both internal and external PACS. Prior imaging findings were credited with answering a clinical question in 98 (23.5%), and a radiographic question in 228 (54.7%) of 417 reports. The process negated the need for follow-up diagnostic testing recommendations in 119 (28.5%) instances, leading to additional imaging recommendations in 19 (4.6%). ConclusionData obtained in this study suggest that comparing current and previous imaging may help reduce unnecessary additional imaging or follow-up diagnostic testing recommendations. Prior imaging review may also facilitate diagnostic confidence and interpretation clarity.

The Role of Chorionicity in Placenta-Related Disorders

Introduction: Twin pregnancy is associated with higher rates of maternal morbidities including gestational diabetes and hypertension. Dichorionic twins are believed to have greater placental mass. Our objective was to study the incidence of two placenta-related disorders: gestational diabetes and hypertension, in dichorionic versus monochorionic twin pregnancies. Methods: Patients’ data of all consecutive twin pregnancies over a period of 12 years were collected from medical records. Data on chorionicity were retrieved from 1st trimester ultrasound reports. Maternal complications including gestational hypertension and diabetes were collected, and incidence was compared between dichorionic and monochorionic twin pregnancies. Records lacking chorionicity data and cases with pregestational diabetes were excluded. Results: A total of 960 twin pregnancies, 121 monochorionic and 839 dichorionic, were included. Average maternal age did not differ significantly between the groups. The median gestational age at delivery was 36.0 weeks in monochorionic and 36.7 in dichorionic twins. Primiparity (40.4% vs. 23.1%, p < 0.001) and the rate of infertility treatments (51.5% vs. 7.4%, p < 0.001) were both more common in the dichorionic group. The incidence of gestational hypertension disorders was 14% in monochorionic versus 11% in dichorionic twins (p = 0.36). Gestational diabetes was more common in dichorionic compared to monochorionic twins (9.4% and 2.5%, respectively); however, logistic multivariate analysis showed that gestational diabetes was highly correlated with maternal age (p < 0.001) and infertility treatments (p < 0.001) but not with chorionicity (p = 0.136). Conclusion: Our results may imply that greater placental mass does not increase the risk for gestational hypertension and diabetes. This might support the role of additional multiple maternal factors associated with these complications.

Sonographic Characteristics of ≥7 mm Gallbladder Polyps: A Retrospective Analysis.

To describe the sonographic characteristics of gallbladder polyps measuring ≥7 mm focusing on echogenicity, correlate with surgical pathology when available, and assess stability in size over time. This retrospective study used a natural language processing application to screen ultrasound (US) reports between January 1, 2012, and December 31, 2020, that contained the words "gallbladder polyp" or "polyps." Reports were reviewed to identify polyps ≥7 mm. The most hyperechoic components of the polyps were compared to the adjacent inner wall of the gallbladder and categorized as more echogenic, isoechoic, and less echogenic. Other sonographic characteristics such as heterogeneity, sessile configuration, vascularity, multiplicity, presence of gallstones, and wall thickening were recorded. Surgical pathology reports were reviewed when available. Polyps in nonsurgical patients with ≥48-month US follow-up and ≤1 mm/year growth rate were characterized as benign. Clinical outcomes were followed until December 31, 2023. Review of 4897 reports yielded 550 reports in 450 patients with polyps ≥7 mm. Surgical pathology reports were available in 22.0% (99/450) of patients; 96 (97%) had non-neoplastic etiologies and 3 (3.0%) neoplastic. There were no malignancies. All of the neoplastic polyps and 56.1% (87/155) of non-neoplastic polyps had components more echogenic than the adjacent inner wall. There were no deaths related to the polyps. The vast majority of gallbladder polyps are benign. In our large series of gallbladder polyps, we found that increased echogenicity is a nonspecific feature, found in slightly more than half of benign but also in all neoplastic polyps. Our findings support current Society of Radiologists in Ultrasound consensus guidelines.

Risk of malignancy in thyroid nodules with increased 11C-Choline uptake detected incidentally on PET/CT: A diagnostic accuracy study.

The purpose was to evaluate the pathological nature of focal thyroid uptake seen in 11C-Choline PET/CT performed for prostate cancer. The study was IRB-approved. All 11C-Choline PET/CT exam reports for studies performed between January 01, 2018, and July 30, 2021, in male patients with prostate cancer in our institution were retrospectively reviewed. Exams with "focal thyroid uptake" on their final report were selected. Patients with surgery or ablation in the thyroid prior to the PET/CT, proven parathyroid adenomas or absent thyroid ultrasound were excluded. Repeated PET/CT exams of same patient were excluded. PET images were analyzed visually and semi-quantitatively by measuring the maximum standardized uptake value (SUVmax) of the focal thyroid uptake. Available thyroid ultrasound images, cytology and pathology reports were reviewed. Statistical analyses were performed. Out of 10,047 sequential 11C-Choline PET/CT studies, 318 reports included "focal thyroid uptake." About 128 of these studies were repeat exams and were excluded. Additional 87 patients were excluded, because the uptake was determined to be adjacent, rather than confined to the thyroid gland. Out of the remaining 103 patients, 74 patients had focal thyroid uptake and concurrent thyroid sonographic evaluation. Out of the 74 focal uptakes evaluated with ultrasound, 21 were presumed benign thyroid nodules based on the ultrasound and 53 had further evaluation with biopsy. Sixty three nodules were benign (21 presumed benign on ultrasound and 42 cytology or surgical pathology-proven), 9 nodules were malignant and 2 remained indeterminate. There was no significant difference between the SUVs of the benign and malignant groups (P > .3). In this retrospective study of patients with prostate cancer who underwent 11C-Choline PET/CT, we identified a group of patients who underwent thyroid ultrasound for incidental finding of focal 11C-Choline thyroid uptake. Incidence of malignancy in this group was 12%. Therefore, further investigation with ultrasound and possibly ultrasound-guided biopsy may be warranted when a choline avid thyroid nodule is found incidentally on choline PET.

Serum arsenic augments gallstone risk in Henan rural cohort with multiple metal exposure

BackgroundPeople are exposed to metals in various ways during their daily lives. However, the association between metal exposure and gallstones remains unclear. ObjectivesTo investigate the relationship between serum elemental concentrations and the risk of gallstones. MethodsParticipants (n = 4204) were drawn from the Henan Rural Cohort. Gallstone diagnosis was based on abdominal ultrasound reports during follow-up. Baseline serum elemental concentrations were measured using inductively coupled plasma mass spectrometry. The relationship between serum elemental levels and gallstones was evaluated using robust Poisson regression, restricted cubic spline (RCS), quantile g-computation (Qgcomp), grouped weighted quantile sum (GWQS) and Bayesian kernel machine regression (BKMR). Results121 individuals were diagnosed with gallstone (incidence rate of 2.88 %). In robust Poisson regression, after adjusting for confounding factors, the highest quartile of arsenic concentration compared to the lowest quartile had a 1.90 times higher relative risk (RR) [95 % confidence interval (CI): 1.05, 3.44]. Conversely, the highest quartile of zinc concentration compared to the lowest quartile had a 0.50 times lower RR (95 % CI: 0.28, 0.89). RCS showed an approximately “S”-shaped nonlinear relationship between serum arsenic levels and gallstones, with increasing arsenic concentration leading to a higher risk of gallstones; however, the risk plateaued when arsenic concentration exceeded 0.62 μg/L. Both the Qgcomp and GWQS indicated that arsenic plays a significant role in increasing the risk of gallstones, whereas zinc plays a significant role in reducing the risk of gallstones. BKMR showed that raising arsenic exposure from the 25th to the 75th percentile increased the risk of gallstones, while raising serum zinc concentration reduced it. ConclusionsHigher serum arsenic concentration increases the risk of gallstones, whereas higher zinc concentration may reduce the risk. Effective prevention of gallstones may require further reduction of arsenic exposure and appropriate increases in zinc intake.

Preliminary experiments on interpretable ChatGPT-assisted diagnosis for breast ultrasound radiologists.

Ultrasound is essential for detecting breast lesions. The American College of Radiology's Breast Imaging Reporting and Data System (BI-RADS) classification system is widely used, but its subjectivity can lead to inconsistency in diagnostic outcomes. Artificial intelligence (AI) models, such as ChatGPT-3.5, may potentially enhance diagnostic accuracy and efficiency in medical settings. This study aimed to assess the utility of the ChatGPT-3.5 model in generating BI-RADS classifications for breast ultrasound reports and its ability to replicate the "chain of thought" (CoT) in clinical decision-making to improve model interpretability. Breast ultrasound reports were collected, and ChatGPT-3.5 was used to generate diagnoses and treatment plans. We evaluated GPT-4's performance by comparing its generated reports to those from doctors with various levels of experience. We also conducted a Turing test and a consistency analysis. To enhance the interpretability of the model, we applied the CoT method to deconstruct the decision-making chain of the GPT model. A total of 131 patients were evaluated, with 57 doctors participating in the experiment. ChatGPT-3.5 showed promising performance in structure and organization (S&O), professional terminology and expression (PTE), treatment recommendations (TR), and clarity and comprehensibility (C&C). However, improvements are needed in BI-RADS classification, malignancy diagnosis (MD), likelihood of being written by a physician (LWBP), and ultrasound doctor artificial intelligence acceptance (UDAIA). Turing test results indicated that AI-generated reports convincingly resembled human-authored reports. Reproducibility experiments displayed consistent performance. Erroneous report analysis revealed issues related to incorrect diagnosis, inconsistencies, and overdiagnosis. The CoT investigation supports the potential of ChatGPT to replicate the clinical decision-making process and offers insights into AI interpretability. The ChatGPT-3.5 model holds potential as a valuable tool for assisting in the efficient determination of BI-RADS classifications and enhancing diagnostic performance.

AI-Powered Synthesis of Structured Multimodal Breast Ultrasound Reports Integrating Radiologist Annotations and Deep Learning Analysis.

Breast cancer is the most prevalent cancer among women worldwide. B-mode ultrasound (US) is essential for early detection, offering high sensitivity and specificity without radiation exposure. This study introduces a semi-automatic method to streamline breast US report generation, aiming to reduce the burden on radiologists. Our method synthesizes comprehensive breast US reports by combining the extracted information from radiologists' annotations during routine screenings with the analysis results from deep learning algorithms on multimodal US images. Key modules in our method include image classification using visual features (ICVF), type classification via deep learning (TCDL), and automatic report structuring and compilation (ARSC). Experiments showed that the proposed method reduced the average report generation time to 3.8 min compared to manual processes, even when using relatively low-spec hardware. Generated reports perfectly matched ground truth reports for suspicious masses without a single failure on our evaluation datasets. Additionally, the deep-learning-based algorithm, utilizing DenseNet-121 as its core model, achieved an overall accuracy of 0.865, precision of 0.868, recall of 0.847, F1-score of 0.856, and area under the receiver operating characteristics of 0.92 in classifying tissue stiffness in breast US shear-wave elastography (SWE-mode) images. These improvements not only streamline the report generation process but also allow radiologists to dedicate more time and focus on patient care, ultimately enhancing clinical outcomes and patient satisfaction.

Obstetric outcomes in the expectant management of cesarean scar pregnancy with fetal heart activity: a single-center retrospective cohort study.

Cesarean scar pregnancy (CSP) is a high-risk complication characterized by the implantation of a pregnancy within a cesarean scar resulting from a previous delivery. Currently, clinical indicators guiding the expectant management of patients with CSP are lacking. We thus aimed to evaluate pregnancy and neonatal outcomes among women who underwent expectant CSP management and to investigate whether sonographic signs correlated with obstetric outcomes. We retrospective reviewed the electronic medical records and first-trimester transvaginal ultrasonography reports of consecutive patients diagnosed with CSP in the first trimester at the West China Second University Hospital from January 1, 2010 to December 31, 2022. Pregnancy outcomes (emergency surgery, blood loss, and rescue) and neonatal outcomes (gestational age at delivery, neonatal weight, and Apgar scores) were examined. A binary logistic regression analysis was conducted to identify independent risk factors that could predict severe complications. The final analysis included 54 patients. The mean age of the pregnant women was 34±4 years. Among the 54 patients, 14 (25.9%) did not progress to 20 weeks of gestation. Pregnancy continued beyond 20 weeks in 40 patients, with 37 live births (92.5%) and 3 stillbirths (7.5%). Moreover, 7 (17.5%) and 33 (82.5%) patients delivered before and after 34 weeks, respectively. Placenta accreta spectrum (PAS) and placenta previa were confirmed in 29 (72.5%) and 17 (42.5%) patients, respectively. Hysterectomy, emergency cesarean section, and rescue surgery were performed in 5 (12.5%), 15 (37.5%), and 22 (55.5%) patients, respectively. Patients with a visible niche were significantly more likely to have preterm labor, PAS, placenta previa, low-birth-weight newborns, higher blood loss, intraoperative rescue, blood transfusion, and first-trimester vaginal bleeding than were those without one (all P values <0.05). Our study showed that expectant management of CSP to achieve live birth might be feasible. Patients with a visible niche exhibited worse outcomes, with a higher incidence of severe delivery complications.

2142: Clinical data extraction: Prompts for breast cancer pathology & ultrasound reports

2142: Clinical data extraction: Prompts for breast cancer pathology & ultrasound reports

Association of Insulin Resistance and Ectopic Fat Accumulation with HOMA Indices: A Single-Centre Observational Study

Purpose: The aim of this study was to investigate the relationship between non-alcoholic fatty liver disease (NAFLD), non-alcoholic fatty pancreas (or pancreatic) disease (NAFPD) and HOMA indices in obese patients without a diagnosis of diabetes mellitus, using ultrasound (US) as a common non-invasive diagnostic tool during routine examinations. Methods: In this single-centre, retrospective study, the records of patients who applied to the obesity outpatient clinic in 2023 were reviewed. Digital records were scanned and patients with abdominal ultrasound reports indicating age, gender, body mass index(BMI), fasting plasma glucose, fasting plasma glucose, C-peptide level and degree of pancreatic and hepatic steatosis were included in the study. Patients with known chronic disease or diabetes mellitus and patients with specific drug use were excluded from the study. Homa indices were calculated using fasting plasma glucose and C-peptide levels. Results: A total of 62 patients were included. Body mass index was 39.1, 91% had NAFLD and 82% had NAFPD. There was a significant positive correlation between BMI and NAFLD and NAFPD. HOMA scores revealed a statistically significant impact of NAFLD on insulin resistance (HOMA-IR) and insulin sensitivity (HOMA-S) but not on beta-cell function (HOMA-B). No significant effect of NAFPD on HOMA scores was observed. Conclusion: The findings underline the association between NAFLD and insulin resistance and highlight the metabolic burden of ectopic fat deposition in obese patients. In contrast, there was no significant correlation between NAFPD and either insulin resistance or beta-cell function, suggesting that the metabolic impact of pancreatic steatosis may be different. These findings may help to guide clinical strategies for detecting and treating metabolic disorders in obesity.

Rehabilitation of a neonate with persistent caput succedaneum using cranial remoulding orthosis (helmet): a case report

A second-born male child to nonconsanguineous parents was delivered by normal vaginal delivery following a difficult extraction at term and had a normal transition to postnatal life. At 40 days of life, the baby presented with a soft, boggy swelling in the posterior occipital region, which had been noticed since birth but was increasing in size gradually, and he was born through a difficult normal vaginal delivery. The preliminary diagnosis considered at admission was probable caput succedaneum, with a differential diagnosis of spontaneous CSF leak. Cranial ultrasound reports showed a cystic lesion measuring 37×28 mm in the right parietooccipital region, suggestive of caput succedaneum. So, the baby was aided by our rehabilitation team experts, who provided the cranial remolding orthosis (helmet) to the neonate, which would provide adequate soft, consistent compression. At 3 and a half months of age, the caput subsided, and the baby was doing clinically well with normal developmental milestones and a normal head circumference, appropriate for the age. Thus, the use of the helmet resulted in a significant improvement in the baby's condition, demonstrating the effectiveness of this intervention for treating persistent caput succedaneum. Therefore, this article highlights the importance of early intervention and specialized care in managing cranial deformities in neonates.

Generative Large Language Models in Electronic Health Records for Patient Care Since 2023: A Systematic Review.

Generative Large language models (LLMs) represent a significant advancement in natural language processing, achieving state-of-the-art performance across various tasks. However, their application in clinical settings using real electronic health records (EHRs) is still rare and presents numerous challenges. This study aims to systematically review the use of generative LLMs, and the effectiveness of relevant techniques in patient care-related topics involving EHRs, summarize the challenges faced, and suggest future directions. A Boolean search for peer-reviewed articles was conducted on May 19th, 2024 using PubMed and Web of Science to include research articles published since 2023, which was one month after the release of ChatGPT. The search results were deduplicated. Multiple reviewers, including biomedical informaticians, computer scientists, and a physician, screened the publications for eligibility and conducted data extraction. Only studies utilizing generative LLMs to analyze real EHR data were included. We summarized the use of prompt engineering, fine-tuning, multimodal EHR data, and evaluation matrices. Additionally, we identified current challenges in applying LLMs in clinical settings as reported by the included studies and proposed future directions. The initial search identified 6,328 unique studies, with 76 studies included after eligibility screening. Of these, 67 studies (88.2%) employed zero-shot prompting, five of them reported 100% accuracy on five specific clinical tasks. Nine studies used advanced prompting strategies; four tested these strategies experimentally, finding that prompt engineering improved performance, with one study noting a non-linear relationship between the number of examples in a prompt and performance improvement. Eight studies explored fine-tuning generative LLMs, all reported performance improvements on specific tasks, but three of them noted potential performance degradation after fine-tuning on certain tasks. Only two studies utilized multimodal data, which improved LLM-based decision-making and enabled accurate rare disease diagnosis and prognosis. The studies employed 55 different evaluation metrics for 22 purposes, such as correctness, completeness, and conciseness. Two studies investigated LLM bias, with one detecting no bias and the other finding that male patients received more appropriate clinical decision-making suggestions. Six studies identified hallucinations, such as fabricating patient names in structured thyroid ultrasound reports. Additional challenges included but were not limited to the impersonal tone of LLM consultations, which made patients uncomfortable, and the difficulty patients had in understanding LLM responses. Our review indicates that few studies have employed advanced computational techniques to enhance LLM performance. The diverse evaluation metrics used highlight the need for standardization. LLMs currently cannot replace physicians due to challenges such as bias, hallucinations, and impersonal responses.

Evaluation of combined hepatocellular-cholangiocarcinoma using CEUS LI-RADS: correlation with pathological characteristics.

To explore the factors that influence the contrast-enhanced ultrasound (CEUS) Liver Imaging Reporting and Data System (LI-RADS) classification of combined hepatocellular-cholangiocarcinoma (cHCC-ICC). Between September 2014 to July 2020, the CEUS features of 58 patients with pathologically confirmed cHCC-ICC were retrospectively evaluated and assigned according to the CEUS LI-RADS (version 2017). The pathological characteristics of nodules categorizing as different CEUS LI-RADS categories were compared. Multivariate logistic regression analysis was conducted to explore potential factors that may influence the CEUS LI-RADS classification of cHCC-ICC. According to CEUS LI-RADS, 32.8% (19/58), 63.8% (37/58), and 3.4% (2/58) were categorized as LR-5, LR-M, and LR-TIV, respectively. There was significant difference between the LR-M and LR-5 groups with regard to the pathological grade, nodule size, and HCC/ICC-component ratio of cHCC-ICC. Multivariate logistic regression analysis identified tumor size and the relative proportions of hepatocellular carcinomas (HCC) and intrahepatic cholangiocarcinomas (ICC) components as the independent influencing factors. Tumor size and the relative proportion of HCC and ICC components within the nodule had a significant impact on the CEUS LI-RADS classification of cHCC-ICC.

The impact of stent protrusion into the inferior vena cava or jailing of the contralateral iliac vein on the incidence of contralateral deep vein thrombosis following venous stenting.

Iliac vein stenting is the standard of care for patients with pelvic venous disorders secondary to symptomatic iliac vein outflow obstruction. Venous stents are often extended proximally into the inferior vena cava (IVC) which may result in partial or complete coverage of the contralateral iliac vein. The purpose of this investigation is to determine if extension of iliac vein stents into the IVC results in increased risk of contralateral deep venous thrombosis (DVT). We retrospectively reviewed prospectively collected data from 409 patients who underwent iliac vein stenting at the Center for Vascular Medicine (CVM) from 2019 to 2020. Stent type, covered territories, initial and follow-up consults, ultrasound and operative reports were reviewed to assess for incidence of post-implantation DVT. Patients were stratified into three groups: Iliac vein stents which protruded into the IVC, stents that completely covered the orifice of the contralateral iliac vein and those with no stent protrusion into the IVC. Out of 409 patients, the average age was 53.96 ± 13.40 years with 94 males and 315 females. All stents placed were Venovo stents and all iliac vein lesions were non-thrombotic stenoses. The average follow-up period was 14.35 ± 10.09 months. The most common territories stented were the IVC-LCIV-LEIV (n = , 74%) and the IVC-RCIV-REIV (n = , 26%). Stent protrusion and distance into the IVC in millimeters (mm) was the following: Partial protrusion (n = 314, 77%, 27.6 ± 19.1), jailing of the contralateral iliac vein (n = 78, 19%, 45.9 ± 18.6), no protrusion (n = 16, 4%). The overall DVT rate post-implantation was 0.49% (n = 2). No DVTs ipsilateral to the index stent were identified and both DVTs were contralateral DVTs. A hypercoaguable disorder was reported in 6 patients (1.5%). There were no significant differences in prevalence of contralateral DVT between the three groups. (p = .35). The rate of contralateral DVTs post iliac vein stenting with Nitonol based stents is extremely low. Partial or complete coverage of the contralateral iliac vein via stenting does not result in an increased incidence of contralateral DVT in the short-term. Longer follow up is needed to determine if contralateral DVTs occur after long-term implantation.

Comparative Analysis of Large Language Models in Simplifying Turkish Ultrasound Reports to Enhance Patient Understanding

Objective: To evaluate and compare the abilities of Language Models (LLMs) in simplifying Turkish ultrasound (US) findings for patients. Methods: We assessed the simplification performance of four LLMs: ChatGPT 4, Gemini 1.5 Pro, Claude 3 Opus, and Perplexity, using fifty fictional Turkish US findings. Comparison was based on Ateşman’s Readability Index and word count. Three radiologists rated medical accuracy, consistency, and comprehensibility on a Likert scale from 1 to 5. Statistical tests (Friedman, Wilcoxon, and Spearman correlation) examined differences in LLMs' performance. Results: Gemini 1.5 Pro, ChatGPT-4, and Claude 3 Opus received high Likert scores for medical accuracy, consistency, and comprehensibility (mean: 4.7–4.8). Perplexity scored significantly lower (mean: 4.1, p&lt;0.001). Gemini 1.5 Pro achieved the highest readability score (mean: 61.16), followed by ChatGPT-4 (mean: 58.94) and Claude 3 Opus (mean: 51.16). Perplexity had the lowest readability score (mean: 47.01). Gemini 1.5 Pro and ChatGPT-4 used significantly more words compared to Claude 3 Opus and Perplexity (p&lt;0.001). Linear correlation analysis revealed a positive correlation between word count of fictional US findings and responses generated by Gemini 1.5 Pro (correlation coefficient = 0.38, p&lt;0.05) and ChatGPT-4 (correlation coefficient = 0.43, p&lt;0.001). Conclusion: This study highlights strong potential of LLMs in simplifying Turkish US findings, improving accessibility and clarity for patients. Gemini 1.5 Pro, ChatGPT-4, and Claude 3 Opus performed well, highlighting their effectiveness in healthcare communication. Further research is required to fully understand the integration of LLMs into clinical practice and their influence on patient comprehension and decision-making.