Liver metastases and hepatocellular carcinomas are two of the most common causes of cancer deaths in the world.Radiofrequency ablation(RFA) is a well recognized,effective and minimally invasive means of treating malig...Liver metastases and hepatocellular carcinomas are two of the most common causes of cancer deaths in the world.Radiofrequency ablation(RFA) is a well recognized,effective and minimally invasive means of treating malignant hepatic tumors.This article describes the use of contrast-enhanced 3D ultrasound(CE-3DUS) in the staging,targeting and follow-up of patients with liver tumors undergoing RFA.In particular,its value in the management of large hepatic lesions will be illustrated.Current limitations of CE-3DUS and future developments in the technique will also be discussed.In summary,CE-3DUS is useful in the RFA of liver tumors with improved detection and display of occult lesions and recurrence,in the assessment of lesional geometry and orientation for a more accurate planning and guidance of multiple RFA needle electrodes in large tumors and in the evaluation of residual or recurrent disease within the immediate and/or subsequent follow-up periods.展开更多
Objectives: This study focused on evaluating the value of transvaginal 3D ultrasound for the diagnosis and prognostic assessment of post-cesarean section uterine diverticulum. Materials and Methods: We retrospectively...Objectives: This study focused on evaluating the value of transvaginal 3D ultrasound for the diagnosis and prognostic assessment of post-cesarean section uterine diverticulum. Materials and Methods: We retrospectively analyzed the data from 32 patients with post-cesarean section uterine diverticulum over three recent years. In all patients, transvaginal 3D ultrasound was used to measure the size of the uterine diverticulum and the thickness of the lower uterine segment (LUS) and myometrium. Patients with a LUS with a myometrial thickness under 4 mm underwent resection and repair surgery;those with a LUS with a myometrial thickness over 4 mm underwent hysteroscopic resection. The postoperative sonograms were compared with preoperative images to evaluate the efficacy of various treatments. Results: The mean length, width and depth of the uterine diverticula were 18.30 ± 2.80 mm, 9.14 ± 3.20 mm and 11.49 ± 2.71 mm, respectively. The average LUS myometrial thickness was 3.40 ± 0.80 mm (with a range of 1.6 mm - 6.3 mm). After surgery, two patients’ sonograms still showed diverticula at the post-cesarean section scar, measuring 6 mm × 7 mm × 6 mm and 6 mm × 8 mm × 4 mm. There were significant differences in the size of uterine diverticula between preoperative and postoperative sonograms (P < 0.05) and the effective rate of surgery was 93.75% (30/32). Conclusions: Transvaginal 3D ultrasound is an accurate method for detecting post-cesarean section uterine diverticulum and is helpful for assessing surgical options and prognosis. The LUS myometrial thickness, which is considered as an optional index of post-cesarean section uterine diverticulum, should be measured routinely.展开更多
Background Deep convolutional neural networks have garnered considerable attention in numerous machine learning applications,particularly in visual recognition tasks such as image and video analyses.There is a growing...Background Deep convolutional neural networks have garnered considerable attention in numerous machine learning applications,particularly in visual recognition tasks such as image and video analyses.There is a growing interest in applying this technology to diverse applications in medical image analysis.Automated three dimensional Breast Ultrasound is a vital tool for detecting breast cancer,and computer-assisted diagnosis software,developed based on deep learning,can effectively assist radiologists in diagnosis.However,the network model is prone to overfitting during training,owing to challenges such as insufficient training data.This study attempts to solve the problem caused by small datasets and improve model detection performance.Methods We propose a breast cancer detection framework based on deep learning(a transfer learning method based on cross-organ cancer detection)and a contrastive learning method based on breast imaging reporting and data systems(BI-RADS).Results When using cross organ transfer learning and BIRADS based contrastive learning,the average sensitivity of the model increased by a maximum of 16.05%.Conclusion Our experiments have demonstrated that the parameters and experiences of cross-organ cancer detection can be mutually referenced,and contrastive learning method based on BI-RADS can improve the detection performance of the model.展开更多
Mild adolescent idiopathic scoliosis(AIS),with Cobb<20°,was hypothesized as the right stage to intervene to prevent progression.AIS curve can be categorized into either structural or non-structural depending o...Mild adolescent idiopathic scoliosis(AIS),with Cobb<20°,was hypothesized as the right stage to intervene to prevent progression.AIS curve can be categorized into either structural or non-structural depending on the spine morphology(flexibility).Using X-ray to characterize AIS curves remains the clinical gold standard whilecompromising the risks of radiation exposure.In previous works,3D ultrasound imaging had proved the reli-ability of the coronal spinal curvature measurement This research aimed at developing a mild AIs clasificationscheme through examining spine flexibility using 3D ultrasound imaging.For the preliminary study,90 mild AIS subjects(21 M and 69 F;Age:14.5±1.7 years old;Cobb:18.2±6.4°)underwent both 3D ultrasound and X-ray scanning on the same day.For each case,a clinician measured Cobbs and denoted major curve as ground truth.Bending Asymmetry Index(BAl)was developed to indicate the presence of a possible structural curve.The curve classification was coded to a modfied Lenke classification for mild cases(m-Lenke).The results of 3D ultrasound classification were evaluated with the X-ray.It was shown that 70.1%of the subjects had identical curve clssification results and 72.0%had the correct major curve detection.Lumbar-dominated curves had distinctive performance(p=0.91,r=0.91)against others.The study demonstrated the possibility of a 3D ultrasound-based method for mild AIS curve classification.Thediscrepancies could be partially explained by the limitations of the ultrasound scanning in proximal thoracicreg ion.Subsequent studies will validate the proposed method with a larger cohort.展开更多
Purpose: This review examines the diagnostic value of transvaginal 3D ultrasound image texture analysis for the diagnosis of uterine adhesions. Materials and Methods: The total clinical data of 53 patients with uterin...Purpose: This review examines the diagnostic value of transvaginal 3D ultrasound image texture analysis for the diagnosis of uterine adhesions. Materials and Methods: The total clinical data of 53 patients with uterine adhesions diagnosed by hysteroscopy and the imaging data of transvaginal three-dimensional ultrasound from the Second Affiliated Hospital of Chongqing Medical University from June 2022 to August 2023 were retrospectively analysed. Based on hysteroscopic surgical records, patients were divided into two independent groups: normal endometrium and uterine adhesion sites. The samples were divided into a training set and a test set, and the transvaginal 3D ultrasound was used to outline the region of interest (ROI) and extract texture features for normal endometrium and uterine adhesions based on hysteroscopic surgical recordings, the training set data were feature screened and modelled using lasso regression and cross-validation, and the diagnostic efficacy of the model was assessed by applying the subjects’ operating characteristic (ROC) curves. Results: For each group, 290 texture feature parameters were extracted and three higher values were screened out, and the area under the curve of the constructed ultrasonographic scoring model was 0.658 and 0.720 in the training and test sets, respectively. Conclusion Relative clinical value of transvaginal three-dimensional ultrasound image texture analysis for the diagnosis of uterine adhesions.展开更多
文摘Liver metastases and hepatocellular carcinomas are two of the most common causes of cancer deaths in the world.Radiofrequency ablation(RFA) is a well recognized,effective and minimally invasive means of treating malignant hepatic tumors.This article describes the use of contrast-enhanced 3D ultrasound(CE-3DUS) in the staging,targeting and follow-up of patients with liver tumors undergoing RFA.In particular,its value in the management of large hepatic lesions will be illustrated.Current limitations of CE-3DUS and future developments in the technique will also be discussed.In summary,CE-3DUS is useful in the RFA of liver tumors with improved detection and display of occult lesions and recurrence,in the assessment of lesional geometry and orientation for a more accurate planning and guidance of multiple RFA needle electrodes in large tumors and in the evaluation of residual or recurrent disease within the immediate and/or subsequent follow-up periods.
文摘Objectives: This study focused on evaluating the value of transvaginal 3D ultrasound for the diagnosis and prognostic assessment of post-cesarean section uterine diverticulum. Materials and Methods: We retrospectively analyzed the data from 32 patients with post-cesarean section uterine diverticulum over three recent years. In all patients, transvaginal 3D ultrasound was used to measure the size of the uterine diverticulum and the thickness of the lower uterine segment (LUS) and myometrium. Patients with a LUS with a myometrial thickness under 4 mm underwent resection and repair surgery;those with a LUS with a myometrial thickness over 4 mm underwent hysteroscopic resection. The postoperative sonograms were compared with preoperative images to evaluate the efficacy of various treatments. Results: The mean length, width and depth of the uterine diverticula were 18.30 ± 2.80 mm, 9.14 ± 3.20 mm and 11.49 ± 2.71 mm, respectively. The average LUS myometrial thickness was 3.40 ± 0.80 mm (with a range of 1.6 mm - 6.3 mm). After surgery, two patients’ sonograms still showed diverticula at the post-cesarean section scar, measuring 6 mm × 7 mm × 6 mm and 6 mm × 8 mm × 4 mm. There were significant differences in the size of uterine diverticula between preoperative and postoperative sonograms (P < 0.05) and the effective rate of surgery was 93.75% (30/32). Conclusions: Transvaginal 3D ultrasound is an accurate method for detecting post-cesarean section uterine diverticulum and is helpful for assessing surgical options and prognosis. The LUS myometrial thickness, which is considered as an optional index of post-cesarean section uterine diverticulum, should be measured routinely.
基金Macao Polytechnic University Grant(RP/FCSD-01/2022RP/FCA-05/2022)Science and Technology Development Fund of Macao(0105/2022/A).
文摘Background Deep convolutional neural networks have garnered considerable attention in numerous machine learning applications,particularly in visual recognition tasks such as image and video analyses.There is a growing interest in applying this technology to diverse applications in medical image analysis.Automated three dimensional Breast Ultrasound is a vital tool for detecting breast cancer,and computer-assisted diagnosis software,developed based on deep learning,can effectively assist radiologists in diagnosis.However,the network model is prone to overfitting during training,owing to challenges such as insufficient training data.This study attempts to solve the problem caused by small datasets and improve model detection performance.Methods We propose a breast cancer detection framework based on deep learning(a transfer learning method based on cross-organ cancer detection)and a contrastive learning method based on breast imaging reporting and data systems(BI-RADS).Results When using cross organ transfer learning and BIRADS based contrastive learning,the average sensitivity of the model increased by a maximum of 16.05%.Conclusion Our experiments have demonstrated that the parameters and experiences of cross-organ cancer detection can be mutually referenced,and contrastive learning method based on BI-RADS can improve the detection performance of the model.
基金This study is partially supported by Research Impact Fund of Hong Kong Research Grant Council(R5017-18)Health and Medical Research Fund of the Hong Kong(04152896).
文摘Mild adolescent idiopathic scoliosis(AIS),with Cobb<20°,was hypothesized as the right stage to intervene to prevent progression.AIS curve can be categorized into either structural or non-structural depending on the spine morphology(flexibility).Using X-ray to characterize AIS curves remains the clinical gold standard whilecompromising the risks of radiation exposure.In previous works,3D ultrasound imaging had proved the reli-ability of the coronal spinal curvature measurement This research aimed at developing a mild AIs clasificationscheme through examining spine flexibility using 3D ultrasound imaging.For the preliminary study,90 mild AIS subjects(21 M and 69 F;Age:14.5±1.7 years old;Cobb:18.2±6.4°)underwent both 3D ultrasound and X-ray scanning on the same day.For each case,a clinician measured Cobbs and denoted major curve as ground truth.Bending Asymmetry Index(BAl)was developed to indicate the presence of a possible structural curve.The curve classification was coded to a modfied Lenke classification for mild cases(m-Lenke).The results of 3D ultrasound classification were evaluated with the X-ray.It was shown that 70.1%of the subjects had identical curve clssification results and 72.0%had the correct major curve detection.Lumbar-dominated curves had distinctive performance(p=0.91,r=0.91)against others.The study demonstrated the possibility of a 3D ultrasound-based method for mild AIS curve classification.Thediscrepancies could be partially explained by the limitations of the ultrasound scanning in proximal thoracicreg ion.Subsequent studies will validate the proposed method with a larger cohort.
文摘Purpose: This review examines the diagnostic value of transvaginal 3D ultrasound image texture analysis for the diagnosis of uterine adhesions. Materials and Methods: The total clinical data of 53 patients with uterine adhesions diagnosed by hysteroscopy and the imaging data of transvaginal three-dimensional ultrasound from the Second Affiliated Hospital of Chongqing Medical University from June 2022 to August 2023 were retrospectively analysed. Based on hysteroscopic surgical records, patients were divided into two independent groups: normal endometrium and uterine adhesion sites. The samples were divided into a training set and a test set, and the transvaginal 3D ultrasound was used to outline the region of interest (ROI) and extract texture features for normal endometrium and uterine adhesions based on hysteroscopic surgical recordings, the training set data were feature screened and modelled using lasso regression and cross-validation, and the diagnostic efficacy of the model was assessed by applying the subjects’ operating characteristic (ROC) curves. Results: For each group, 290 texture feature parameters were extracted and three higher values were screened out, and the area under the curve of the constructed ultrasonographic scoring model was 0.658 and 0.720 in the training and test sets, respectively. Conclusion Relative clinical value of transvaginal three-dimensional ultrasound image texture analysis for the diagnosis of uterine adhesions.
