Case report: Ultrasound misdiagnoses atypical parathyroid adenoma as malignant thyroid tumor.

Atypical Parathyroid Adenoma (APA) is a type of tumor that lies somewhere between parathyroid adenoma and parathyroid carcinoma. It often affects adults over the age of 60, and the clinical symptoms are consistent with those of hyperparathyroidism. This condition has a low occurrence, and its ultrasonographic signs are strikingly similar to thyroid malignant tumors, making it easily misdiagnosed. As a result, a case of APA ultrasonography misdiagnosis admitted to our hospital was recorded in order to serve as a reference point for APA diagnosis.

Identifying unstable ramp lesions using ultrasonography.

BACKGROUND: Patients with suspected ramp lesions on magnetic resonance imaging (MRI) or ultrasonography (US) healed and showed no instability based on intraoperative arthroscopic findings. The purpose of this study was to assess the use of US in evaluating ramp lesions preoperatively and intraoperatively. METHODS: Eighty-two knees that underwent anterior cruciate ligament (ACL) reconstruction between January 2022 and June 2023 were included to assess the ramp lesion complication rate and instability using arthroscopic findings. The detection rate of ramp lesions using US at the initial visit and preoperatively was also investigated. The test-retest reliability was assessed using the intraclass correlation coefficient and analyzed using two-way random effects and absolute agreement. The patients were divided into two groups based on the presence or absence of ramp lesions, and these data were compared using Student's t-test. Statistical significance was set at p < 0.05. RESULTS: On ultrasound examination, 90.0% of the cases had a ramp lesion at the initial examination, of which 22.2% were poorly delineated on the day of surgery. In the cases where the ramp lesion was unstable at the time of surgery, it could be delineated using US. In the cases where the ramp lesion was stable, it was difficult to delineate the lesion using US. CONCLUSIONS: Unstable ramp lesions complicating ACL injuries could be detected using US.

Ultrasonic Diagnosis of Congenital Enterocolitis Fistula: A Case Report.

Background: Congenital enterocolic fistula, an abnormal connection between the small intestine and the colon, is a rare condition with the potential for significant complications affecting the patient's quality of life. Case Report: A 2 year and 7 months old girl presented with abdominal pain and diarrhea lasting more than 10 days. The formation of the intestinal fistula was first detected by ultrasound, and the blood flow in the intestinal wall was preliminally analyzed. Surgical exploration revealed a colonic fistula formed by the attachment of the jejunum to the descending colon. Postoperatively, symptoms improved; no secondary infection occurred and the fistula healed well. Conclusion: Congenital colon fistula is rarely reported, and ultrasound is becoming more and more important in its diagnosis. Here, we report a case of congenital colonic fistula diagnosed by ultrasound. Ultrasound can dynamically and in real-time observe the intestinal condition, which is conducive to the early diagnosis and staging of congenital intestinal diseases and the determination of diagnosis and treatment schemes.

Deep learning-assisted ultrasonic diagnosis of cervical lymph node metastasis of thyroid cancer: a retrospective study of 3059 patients.

Objective: This study aimed to develop a deep learning system to identify and differentiate the metastatic cervical lymph nodes (CLNs) of thyroid cancer. Methods: From January 2014 to December 2020, 3059 consecutive patients with suspected with metastatic CLNs of thyroid cancer were retrospectively enrolled in this study. All CLNs were confirmed by fine needle aspiration. The patients were randomly divided into the training (1228 benign and 1284 metastatic CLNs) and test (307 benign and 240 metastatic CLNs) groups. Grayscale ultrasonic images were used to develop and test the performance of the Y-Net deep learning model. We used the Y-Net network model to segment and differentiate the lymph nodes. The Dice coefficient was used to evaluate the segmentation efficiency. Sensitivity, specificity, accuracy, positive predictive value (PPV), and negative predictive value (NPV) were used to evaluate the classification efficiency. Results: In the test set, the median Dice coefficient was 0.832. The sensitivity, specificity, accuracy, PPV, and NPV were 57.25%, 87.08%, 72.03%, 81.87%, and 66.67%, respectively. We also used the Y-Net classified branch to evaluate the classification efficiency of the LNs ultrasonic images. The classification branch model had sensitivity, specificity, accuracy, PPV, and NPV of 84.78%, 80.23%, 82.45%, 79.35%, and 85.61%, respectively. For the original ultrasonic reports, the sensitivity, specificity, accuracy, PPV, and NPV were 95.14%, 34.3%, 64.66%, 59.02%, 87.71%, respectively. The Y-Net model yielded better accuracy than the original ultrasonic reports. Conclusion: The Y-Net model can be useful in assisting sonographers to improve the accuracy of the classification of ultrasound images of metastatic CLNs.

Accuracy of Thoracic Ultrasonography for the Diagnosis of Pediatric Pneumonia: A Systematic Review and Meta-Analysis.

As an emerging imaging technique, thoracic ultrasonography (TUS) is increasingly utilized in the diagnosis of lung diseases in children and newborns, especially in emergency and critical settings. This systematic review aimed to estimate the diagnostic accuracy of TUS in childhood pneumonia. We searched Embase, PubMed, and Web of Science for studies until July 2023 using both TUS and chest radiography (CR) for the diagnosis of pediatric pneumonia. Two researchers independently screened the literature based on the inclusion and exclusion criteria, collected the results, and assessed the risk of bias using the Diagnostic Accuracy Study Quality Assessment (QUADAS) tool. A total of 26 articles met our inclusion criteria and were included in the final analysis, including 22 prospective studies and four retrospective studies. The StataMP 14.0 software was used for the analysis of the study. The overall pooled sensitivity was 0.95 [95% confidence intervals (CI), 0.92-0.97] and the specificity was 0.94 [95% CI, 0.88-0.97], depicting a good diagnostic accuracy. Our results indicated that TUS was an effective imaging modality for detecting pediatric pneumonia. It is a potential alternative to CXR and a follow-up for pediatric pneumonia due to its simplicity, versatility, low cost, and lack of radiation hazards.

The dimethadione-exposed rat fetus: an animal model for the prenatal ultrasound characterization of ventricular septal defect.

BACKGROUND: Ventricular septal defect (VSD) is the most prevalent congenital heart disease (CHD) and is easily misdiagnosed or missed. An appropriate VSD animal model could be used to analyze the ultrasound characteristics and their related pathological bases, and provides the opportunity to further explore the pathogenesis of VSD. Currently, little is known about whether ultrahigh-frequency ultrasound biomicroscopy (UBM) is suitable to diagnose VSD of fetal rats. There is no research on whether a dimethadione (DMO)-induced fetal VSD model is suitable for the observation and analysis of imaging characteristics and the associated pathological basis. METHODS: We used DMO to induce VSD. UBM was used to perform the prenatal ultrasound characterization. With the pathological results used as the gold standard, the ultrasound characteristics and their related pathological bases were analyzed. RESULTS: The incidence of VSD in the DMO group was 42.05% and 39.71% (diagnosed by UBM and pathology, respectively, P > 0.05). The prenatal ultrasound findings and pathological basis of various diseases, including isolated VSD, complex CHD containing VSD, and extracardiac lesions, were detected and discussed. It was discovered that some fetuses showed features of noncompacted ventricular myocardium, and for the first time, clusters of red blood cell traversing the cardiomyocytes. CONCLUSIONS: The DMO-induced VSD model is a low-cost model with a high success rate and is suitable for the observation and analysis of VSD. UBM is suitable for evaluating VSD.

First-Trimester Ultrasound Imaging for Prenatal Assessment of the Extended Cardiovascular System Using the Cardiovascular System Sonographic Evaluation Algorithm (CASSEAL).

INTRODUCTION AND OBJECTIVES: To compare fetal images obtained at the first- and second-trimester ultrasound scan when applying the Cardiovascular System Sonographic Evaluation Algorithm (CASSEAL). METHODS: Using the CASSEAL protocol, nine sequential axial views were acquired in B-mode and color Doppler at the first- and second-trimester ultrasound scans, visualizing the main components of the extended fetal cardiovascular system. Images were compared qualitatively between both trimesters. RESULTS: We obtained comparable images for all the nine axial views described in the CASSEAL protocol, with few differences and limitations. CONCLUSIONS: The CASSEAL protocol is reproducible in the first trimester, and could help in the early detection of fetal cardiovascular abnormalities. It represents a promising additional tool in order to increase the CHD detection rate.

Ultrasound diagnosis of renal cell carcinoma associated with Xp11.2 translocation/TFE3 gene fusion in children and adolescents.

BACKGROUND: Many studies have reported Xp 11.2 translocation renal cancer in radioimaging,but there is little literature on the evaluation of Xp11.2 translocation renal cell carcinoma by ultrasound. OBJECTIVE: To investigate the ultrasonographic features and diagnostic value of renal cell carcinoma associated with Xp11.2 translocation/TFE3 gene fusion in children and adolescents. MATERIALS AND METHODS: The clinical and ultrasonographic data of 10 patients with renal cell carcinoma associated with Xp11.2 translocation/TFE3 gene fusion confirmed by pathology in our hospital were analyzed retrospectively. The age ranged from 3 to 18 years old, including 7 males and 3 females. The tumor location, size, boundary, echo, hemorrhage, cystic change, calcification, blood flow, lymph node status and metastasis were mainly observed, and the results were compared with the pathological results. RESULTS: There were 10 masses in 10 cases of renal cell carcinoma associated with Xp11.2 translocation/TFE3 gene fusion, including 4 in the right kidney and 6 in the left kidney; the maximum diameter line is 5-23cm; 9 cases had clear mass boundary (90%); 9 masses (90%) showed mixed cystic and solid masses with high echo of solid components, and 1 mass (10%) showed huge multilocular cystic mass with multiple septations; necrosis and cystic changes were seen in all 10 masses (100%); calcification in 5 masses (50%); blood flow signals were seen in the solid components of the mass (100%). CONCLUSION: Renal cell carcinoma associated with Xp11.2 translocation/TFE3 gene fusion in children and adolescents are mostly large cystic and solid mixed echo masses, with high echo of solid components, and often accompanied by cystic changes and calcification. Its ultrasonic manifestations have certain characteristics. Color Doppler ultrasound has a certain diagnostic value for this disease.

Predictive value of ultrasonography in polyneuropathy diagnosis: electrophysiological and ultrasonographic analysis.

Peripheral neuropathy may cause serious complications such as foot ulcers and Charcot joint which can prevent by early diagnosis. We aimed to analyze the diagnostic value of ultrasonographic measurements of nerves and muscles in distal symmetric axonal polyneuropathy (DSAP). Study included 51 DSAP patients and 51 controls. Nerve conduction studies were performed. Median, ulnar, tibial, superficial peroneal, and sural nerves and the abductor pollicis brevis (APB), abductor digiti minimi (ADM), first dorsal interosseous (FDI), extensor digitorum brevis (EDB), abductor hallucis (AH) and tibialis anterior (TA) muscles were evaluated with ultrasound. The Toronto clinical scoring system (TCSS) was used to assess the severity of neuropathy. The median, ulnar, and tibial nerve cross-sectional areas (CSA) were higher in the DSAP group (p = 0.025, p = 0.011, p < 0.001 respectively) while superficial peroneal and sural nerve CSAs were not differed. Only AH and EDB ultrasonographic findings from the muscles differed between the two groups. Effect of diabetes and DSAP on sonographic findings were assessed with two-way ANOVA. Results indicated that only DSAP had a significant effect on sonographic nerve and muscle examination. The area under the ROC curve was 0.831 ± 0.042 for tibial nerve CSA (p < 0.001) with a cut-off value of 15.5 mm2 (sensitivity 74% and specificity 83%). Median, ulnar and tibial nerve CSAs were found to be larger in polyneuropathy patients and they were associated with the clinical and electrophysiological severity of polyneuropathy. ROC analysis showed that tibial nerve CSA may have a predictive value in the diagnosis of DSAP.

The diagnostic accuracy of ultrasonography over manual aspiration for gastric reserve volume estimation in critically ill patients.

BACKGROUND: Although gastric reserve volume (GRV) is a surrogate marker of gastrointestinal dysfunction and feeding intolerance, there is ambiguity in its estimation due to problems associated with its measurement. Introduction of point-of-care ultrasound as a tool for anesthetists kindled interest in its use for GRV estimation. METHODS: In this prospective observational study, we recruited 57 critically ill patients and analyzed 586 samples of GRV obtained by both ultrasonography (USG) and manual aspiration. RESULTS: The analysis showed that USG-guided GRV was significantly correlated (r=0.788, P<0.001) and in positive agreement with manual aspiration based on Bland-Altman plot, with a mean difference of 8.50±14.84 (95% confidence interval, 7.389-9.798). The upper and lower limits of agreement were 37.7 and -20.5, respectively, within the ±1.96 standard deviation (P<0.001). The respective sensitivity and positive predictive value, specificity and negative predictive value, and area under the curve of USG for feeding intolerance were 66.67%, 98.15%, and 0.82%, with 96.49% diagnostic accuracy. CONCLUSIONS: Ultrasonographic estimation of GRV was positively, significantly correlated and in agreement with the manual aspiration method and estimated feeding intolerance earlier. Routine use of gastric USG could avoid clinical situations where feeding status is unclear and there is high risk of aspiration and could become a standard practice of critical care.

A complementary comment on primary hepatic angiosarcoma: A case report.

BACKGROUND: This article examines primary hepatic angiosarcoma (PHA) and fat-poor angiomyolipoma (AML), two uncommon vascular cancers. Clinical decisions in these situations are frequently aided by pathology reports and imaging techniques. Uncommon malignant tumors of the vascular endothelium include PHA. Another diagnosis that should not be overlooked when employing contrast-enhanced MR and contrast-enhanced computed tomography (CT) imaging techniques is fat-poor AML, one of the uncommon vascular tumors of the liver. In both conditions, biopsy is the primary means of diagnosis. CASE SUMMARY: In our article, besides the diagnosis of PHA, fat-poor AML, one of the other rare vascular tumors of the liver, is mentioned. In the case, a 50-year-old female patient with VHL Syndrome was admitted to our hospital with nonspecific lesions such as right upper quadrant pain, weight loss, and nausea. Abdominal ultrasonography (US) revealed a hypoechoic heterogeneous lesion with occasional faint contours. In computed tomography, it was observed as a hyperdense nodular lesion in segment 4. Magnetic resonance imaging (MRI) revealed that the lesion did not contain fat. In connection with the known history of VHL Syndrome, we first evaluated the possibility of AML. Thereupon, a histopathological sample was taken and the diagnosis was made as fat-poor AML with 5% fat content. CONCLUSION: In conclusion, PHA in our case report and fat-poor AML in our clinic are two uncommon liver vascular malignancies with comparable incidences. Important imaging techniques like contrast-enhanced US (CEUS), CECT, and CEMRI give us substantial advantages in both cases. However, a biopsy is used to provide the final diagnosis.

Value Of Anteroposterior-To-Transverse Ratio in Ultrasonic Diagnosis of Thyroid Nodule in Different Locations.

PURPOSE: To investigate the value of anteroposterior-to-transverse ratio (ATR) and the effect on features of nodules in ultrasound (US) diagnosis of thyroid nodules in different locations.  Methods: The nodules were divided into three groups according to the different nodule location: isthmus group; upper and lower poles of bilobed thyroid group; and the middle of the bilobed thyroid group. The diameters of the nodules were recorded, and ATR of the nodule was calculated on the transverse and longitudinal sections. The transverse and the longitudinal sections of ATR of thyroid nodules in different groups were compared.  Result: The transverse section of ATR was significantly different among the three groups (p = 0.001). In addition, there are significant differences in many US features among three groups, including nodule composition, thyroid parenchyma, morphology, echogenicity, shape, calcifications, vascularity, nodule ACR TI-RADS and histopathologic (all p < 0.05). In the group of upper and lower poles of bilobed thyroid, significant difference was found between the transverse and the longitudinal section of ATR (p = 0.019). The cut-off values of transverse section and longitudinal section of ATR were 0.967 and 0.750, respectively.  Conclusion: The transverse section of ATR at different location of thyroid may be a predictor for malignancy with clinical diagnostic significance.

Non-mass Breast Lesions: Could Multimodal Ultrasound Imaging Be Helpful for Their Diagnosis?

Objective: To develop a prediction model for discriminating malignant from benign breast non-mass-like lesions (NMLs) using conventional ultrasound (US), strain elastography (SE) of US elastography and contrast-enhanced ultrasound (CEUS). Methods: A total of 101 NMLs from 100 patients detected by conventional US were enrolled in this retrospective study. The characteristics of NMLs in conventional US, SE and CEUS were compared between malignant and benign NMLs. Histopathological results were used as the reference standard. Binary logistic regression analysis was performed to identify the independent risk factors. A multimodal method to evaluate NMLs based on logistic regression was developed. The diagnostic performance of conventional US, US + SE, US + CEUS and the combination of these modalities was evaluated and compared. Results: Among the 101 lesions, 50 (49.5%) were benign and 51 (50.5%) were malignant. Age ≥45 y, microcalcifications in the lesion, elasticity score >3, earlier enhancement time and hyper-enhancement were independent diagnostic indicators included to establish the multimodal prediction method. The area under the receiver operating characteristic curve (AUC) of US + SE + CEUS was significantly higher than that of US (p < 0.0001) and US + SE (p < 0.0001), but there was no significant difference between the AUC of US + SE + CEUS and the AUC of US + CEUS (p = 0.216). Conclusion: US + SE + CEUS and US + CEUS could significantly improve the diagnostic efficiency and accuracy of conventional US in the diagnosis of NMLs.

Primary hepatic angiosarcoma: A case report.

BACKGROUND: Primary hepatic angiosarcoma (PHA) is a rare malignant tumor of the vascular endothelium. Clinical manifestations and laboratory and imaging examinations often lack specificity for PHA. We report a case of PHA, and describe the ultrasound characteristics and characteristic changes in laboratory values associated with PHA. CASE SUMMARY: A 75-year-old woman presented with right upper quadrant abdominal pain for half a month. Magnetic resonance imaging (MRI) at a local hospital revealed multiple liver space-occupying lesions, and she was admitted to our hospital for further diagnosis. Contrast-enhanced ultrasound (CEUS) revealed multiple slightly hyperechoic nodules in the liver, which were suspected to be of malignant vascular origin. Contrast-enhanced computed tomography revealed multiple low-density nodules in the liver, considered to be metastatic hematopoietic malignancies. Contrast-enhanced MRI showed that the multiple liver nodules shared features with infectious lesions. Laboratory examination revealed normal alpha-fetoprotein levels, slightly increased other liver enzymes, decreased platelets, and significantly increased D-dimer levels. Liver biopsy and histopathology confirmed the presence of PHA. CONCLUSION: CEUS can provide valuable clues for the diagnosis of PHA and greatly improve the success rate of puncture biopsy.

Artificial intelligence for identification of focal lesions in intraoperative liver ultrasonography.

PURPOSE: Intraoperative ultrasonography (IOUS) of the liver is a crucial adjunct in every liver resection and may significantly impact intraoperative surgical decisions. However, IOUS is highly operator dependent and has a steep learning curve. We describe the design and assessment of an artificial intelligence (AI) system to identify focal liver lesions in IOUS. METHODS: IOUS images were collected during liver resections performed between November 2020 and November 2021. The images were labeled by radiologists and surgeons as normal liver tissue versus images that contain liver lesions. A convolutional neural network (CNN) was trained and tested to classify images based on the labeling. Algorithm performance was tested in terms of area under the curves (AUCs), accuracy, sensitivity, specificity, F1 score, positive predictive value, and negative predictive value. RESULTS: Overall, the dataset included 5043 IOUS images from 16 patients. Of these, 2576 were labeled as normal liver tissue and 2467 as containing focal liver lesions. Training and testing image sets were taken from different patients. Network performance area under the curve (AUC) was 80.2 ± 2.9%, and the overall classification accuracy was 74.6% ± 3.1%. For maximal sensitivity of 99%, the classification specificity is 36.4 ± 9.4%. CONCLUSIONS: This study provides for the first time a proof of concept for the use of AI in IOUS and show that high accuracy can be achieved. Further studies using high volume data are warranted to increase accuracy and differentiate between lesion types.

Placenta Accreta Spectrum Prenatal Diagnosis Performance: Are Ultrasound False-positive Results Acceptable in Limited-resources Settings? / Desempenho do diagnóstico pré-natal do espectro da placenta acreta: Os resultados falso-positivos do ultrassom são aceitáveis em ambientes com recursos limitados?

Abstract Objective The immediate referral of patients with risk factors for placenta accreta spectrum (PAS) to specialized centers is recommended, thus favoring an early diagnosis and an interdisciplinary management. However, diagnostic errors are frequent, even in referral centers (RCs). We sought to evaluate the performance of the prenatal diagnosis for PAS in a Latin American hospital. Methods A retrospective descriptive study including patients referred due to the suspicion of PAS was conducted. Data from the prenatal imaging studies were compared with the final diagnoses (intraoperative and/or histological). Results A total of 162 patients were included in the present study. The median gestational age at the time of the first PAS suspicious ultrasound was 29 weeks, but patients arrived at the PAS RC at 34 weeks. The frequency of false-positive results at referring hospitals was 68.5%. Sixty-nine patients underwent surgery based on the suspicion of PAS at 35 weeks, and there was a 28.9% false-positive rate at the RC. In 93 patients, the diagnosis of PAS was ruled out at the RC, with a 2.1% false-negative frequency. Conclusion The prenatal diagnosis of PAS is better at the RC. However, even in these centers, false-positive results are common; therefore, the intraoperative confirmation of the diagnosis of PAS is essential.

Resumo Objetivo Recomenda-se o encaminhamento imediato de pacientes com fatores de risco para espectro placentário acreta (PAS, na sigla em inglês) para centros especializados, favorecendo assim o diagnóstico precoce e o manejo interdisciplinar. No entanto, erros diagnósticos são frequentes, mesmo em centros de referência (CRs). Buscou-se avaliar o desempenho do diagnóstico pré-natal para PAS em um hospital latino-americano. Métodos Um estudo descritivo retrospectivo incluindo pacientes encaminhados por suspeita de SAP foi realizado. Os dados dos exames de imagem do pré-natal foram comparados com os diagnósticos finais (intraoperatórios e/ou histológicos). Resultados Foram incluídos 162 pacientes no presente estudo. A idade gestacional mediana no momento da primeira ultrassonografia suspeita de PAS foi de 29 semanas, mas as pacientes chegaram ao CR de PAS com 34 semanas. A frequência de resultados falso-positivos nos hospitais de referência foi de 68,5%. Sessenta e nove pacientes foram operadas com base na suspeita de PAS com 35 semanas e houve 28,9% de falso-positivos no CR. Em 93 pacientes, o diagnóstico de PAS foi descartado no CR, com frequência de falso-negativos de 2,1%. Conclusão O diagnóstico pré-natal de PAS é melhor no CR. Entretanto, mesmo nestes centros, resultados falso-positivos são comuns; portanto, a confirmação intraoperatória do diagnóstico de SAP é essencial.

High-resolution ultrasonography of gingival biomarkers for periodontal diagnosis in healthy and diseased subjects.

OBJECTIVE: To determine the capacity of ultrasonographic image-based measurements of gingival height and alveolar bone level for monitoring periodontal health and disease. METHODS: Sixteen subjects were recruited from patients scheduled to receive dental care and classified as periodontally healthy (n = 10) or diseased (n = 6) according to clinical guidelines. A 40-MHz ultrasound system was used to measure gingival recession, gingival height, alveolar bone level, and gingival thickness from 66 teeth for comparison to probing measurements of pocket depth and clinical attachment level. Interexaminer variability and comparison between ultrasound measurements and probing measurements was performed via Bland-Altman analysis. RESULTS: Gingival recession and its risk in non-recessed patients could be determined via measurement of the supra- and subgingival cementoenamel junction relative to the gingival margin. Interexaminer bias for ultrasound image analysis was negligible (<0.10 mm) for imaged gingival height (iGH) and 0.45 mm for imaged alveolar bone level (iABL). Diseased subjects had significantly higher imaging measurements (iGH, iABL) and clinical measurements (probing pocket depth, clinical attachment level) than healthy subjects (p < 0.05). Subtraction of the average biologic width from iGH resulted in 83% agreement (≤1 mm difference) between iGH and probing pocket depth measurements. CONCLUSIONS: Ultrasonography has an equivalent diagnostic capacity as gold-standard physical probing for periodontal metrics while offering more detailed anatomical information.

Prenatal prediction of neonatal respiratory morbidity: a radiomics method based on imbalanced few-shot fetal lung ultrasound images.

BACKGROUND: To develop a non-invasive method for the prenatal prediction of neonatal respiratory morbidity (NRM) by a novel radiomics method based on imbalanced few-shot fetal lung ultrasound images. METHODS: A total of 210 fetal lung ultrasound images were enrolled in this study, including 159 normal newborns and 51 NRM newborns. Fetal lungs were delineated as the region of interest (ROI), where radiomics features were designed and extracted. Integrating radiomics features selected and two clinical features, including gestational age and gestational diabetes mellitus, the prediction model was developed and evaluated. The modelling methods used were data augmentation, cost-sensitive learning, and ensemble learning. Furthermore, two methods, which embed data balancing into ensemble learning, were employed to address the problems of imbalance and few-shot simultaneously. RESULTS: Our model achieved sensitivity values of 0.82, specificity values of 0.84, balanced accuracy values of 0.83 and area under the curve values of 0.87 in the test set. The radiomics features extracted from the ROIs at different locations within the lung region achieved similar classification performance outcomes. CONCLUSION: The feature set we designed can efficiently and robustly describe fetal lungs for NRM prediction. RUSBoost shows excellent performance compared to state-of-the-art classifiers on the imbalanced few-shot dataset. The diagnostic efficacy of the model we developed is similar to that of several previous reports of amniocentesis and can serve as a non-invasive, precise evaluation tool for NRM prediction.