scholarly journals Development of a fully automatic deep learning system for L3 selection and body composition assessment on computed tomography

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Jiyeon Ha ◽  
Taeyong Park ◽  
Hong-Kyu Kim ◽  
Youngbin Shin ◽  
Yousun Ko ◽  
...  
Keyword(s):  
Computed Tomography ◽  
Anatomic Variation ◽  
External Validation ◽  
Abdominal Muscle ◽  
Anatomic Variations ◽  
Success Rates ◽  
Technical Success ◽  
Segmentation Accuracy ◽  
Distance Difference ◽  
Fully Automatic

AbstractAs sarcopenia research has been gaining emphasis, the need for quantification of abdominal muscle on computed tomography (CT) is increasing. Thus, a fully automated system to select L3 slice and segment muscle in an end-to-end manner is demanded. We aimed to develop a deep learning model (DLM) to select the L3 slice with consideration of anatomic variations and to segment cross-sectional areas (CSAs) of abdominal muscle and fat. Our DLM, named L3SEG-net, was composed of a YOLOv3-based algorithm for selecting the L3 slice and a fully convolutional network (FCN)-based algorithm for segmentation. The YOLOv3-based algorithm was developed via supervised learning using a training dataset (n = 922), and the FCN-based algorithm was transferred from prior work. Our L3SEG-net was validated with internal (n = 496) and external validation (n = 586) datasets. Ground truth L3 level CT slice and anatomic variation were identified by a board-certified radiologist. L3 slice selection accuracy was evaluated by the distance difference between ground truths and DLM-derived results. Technical success for L3 slice selection was defined when the distance difference was < 10 mm. Overall segmentation accuracy was evaluated by CSA error and DSC value. The influence of anatomic variations on DLM performance was evaluated. In the internal and external validation datasets, the accuracy of automatic L3 slice selection was high, with mean distance differences of 3.7 ± 8.4 mm and 4.1 ± 8.3 mm, respectively, and with technical success rates of 93.1% and 92.3%, respectively. However, in the subgroup analysis of anatomic variations, the L3 slice selection accuracy decreased, with distance differences of 12.4 ± 15.4 mm and 12.1 ± 14.6 mm, respectively, and with technical success rates of 67.2% and 67.9%, respectively. The overall segmentation accuracy of abdominal muscle areas was excellent regardless of anatomic variation, with CSA errors of 1.38–3.10 cm2. A fully automatic system was developed for the selection of an exact axial CT slice at the L3 vertebral level and the segmentation of abdominal muscle areas.

A Fully Automatic Deep Learning System for L3 Slice Selection and Body Composition Assessment on Abdominal Computed Tomography. (Preprint)

2021 ◽  
Author(s):  
Jiyeon Ha ◽  
Taeyong Park ◽  
Hong-Kyu Kim ◽  
Youngbin Shin ◽  
Yousun Ko ◽  
...  
Keyword(s):  
Computed Tomography ◽  
Deep Learning ◽  
External Validation ◽  
Abdominal Muscle ◽  
Anatomic Variations ◽  
Success Rates ◽  
Technical Success ◽  
Segmentation Accuracy ◽  
Distance Difference ◽  
Fully Automatic

BACKGROUND As sarcopenia research has been gaining emphasis, the need for quantification of abdominal muscle on computed tomography (CT) is increasing. Thus, a fully automated system to select L3 slice and segment muscle in an end-to-end manner is demanding. OBJECTIVE We aimed to develop a deep learning model (DLM) to select the L3 slice with consideration of anatomic variations and to segment cross-sectional areas (CSAs) of abdominal muscle and fat. METHODS Our DLM, named L3SEG-net, was composed of a YOLOv3-based algorithm for selecting the L3 slice and a fully convolutional network (FCN)-based algorithm for segmentation. The YOLOv3-based algorithm was developed via supervised learning using a training dataset (n=922), and the FCN-based algorithm was transferred from prior work. Our L3SEG-net was validated with internal (n=496) and external validation (n=586) datasets. L3 slice selection accuracy was evaluated by the distance difference between ground truths and DLM-derived results. Technical success for L3 slice selection was defined when the distance difference was <10 mm. Overall segmentation accuracy was evaluated by CSA error. The influence of anatomic variations on DLM performance was evaluated. RESULTS In the internal and external validation datasets, the accuracy of automatic L3 slice selection was high, with mean distance differences of 3.7±8.4 mm and 4.1±8.3 mm, respectively, and with technical success rates of 93.1% and 92.3%, respectively. However, in the subgroup analysis of anatomic variations, the L3 slice selection accuracy decreased, with distance differences of 12.4±15.4 mm and 12.1±14.6 mm, respectively, and with technical success rates of 67.2% and 67.9%, respectively. The overall segmentation accuracy of abdominal muscle areas was excellent regardless of anatomic variation, with the CSA errors of 1.38–3.10 cm2. CONCLUSIONS A fully automatic system was developed for the selection of an exact axial CT slice at the L3 vertebral level and the segmentation of abdominal muscle areas.

scholarly journals A fully automatic deep learning system for L3 slice selection and body composition assessment on abdominal computed tomography. 

2021 ◽  
Author(s):  
Jiyeon Ha ◽  
Taeyong Park ◽  
Hong-Kyu Kim ◽  
Youngbin Shin ◽  
Yousun Ko ◽  
...  
Keyword(s):  
Computed Tomography ◽  
Deep Learning ◽  
External Validation ◽  
Abdominal Muscle ◽  
Anatomic Variations ◽  
Success Rates ◽  
Technical Success ◽  
Segmentation Accuracy ◽  
Distance Difference ◽  
Fully Automatic

Abstract Background and aims: As sarcopenia research has been gaining emphasis, the need for quantification of abdominal muscle on computed tomography (CT) is increasing. Thus, a fully automated system to select L3 slice and segment muscle in an end-to-end manner is demanded. We aimed to develop a deep learning model (DLM) to select the L3 slice with consideration of anatomic variations and to segment cross-sectional areas (CSAs) of abdominal muscle and fat. Methods: Our DLM, named L3SEG-net, was composed of a YOLOv3-based algorithm for selecting the L3 slice and a fully convolutional network (FCN)-based algorithm for segmentation. The YOLOv3-based algorithm was developed via supervised learning using a training dataset (n=922), and the FCN-based algorithm was transferred from prior work. Our L3SEG-net was validated with internal (n=496) and external validation (n=586) datasets. L3 slice selection accuracy was evaluated by the distance difference between ground truths and DLM-derived results. Technical success for L3 slice selection was defined when the distance difference was <10 mm. Overall segmentation accuracy was evaluated by CSA error. The influence of anatomic variations on DLM performance was evaluated.Results: In the internal and external validation datasets, the accuracy of automatic L3 slice selection was high, with mean distance differences of 3.7±8.4 mm and 4.1±8.3 mm, respectively, and with technical success rates of 93.1% and 92.3%, respectively. However, in the subgroup analysis of anatomic variations, the L3 slice selection accuracy decreased, with distance differences of 12.4±15.4 mm and 12.1±14.6 mm, respectively, and with technical success rates of 67.2% and 67.9%, respectively. The overall segmentation accuracy of abdominal muscle areas was excellent regardless of anatomic variation, with the CSA errors of 1.38–3.10 cm2.Conclusions: A fully automatic system was developed for the selection of an exact axial CT slice at the L3 vertebral level and the segmentation of abdominal muscle areas.

scholarly journals Concha Bullosa in Paradoxical Middle Turbinate: A New Variation

2016 ◽  
Vol 9 (3) ◽  
pp. 141-142 ◽  
Author(s):  
Mohammad W El-Anwar ◽  
Ahmed I Ali
Keyword(s):  
Computed Tomography ◽  
Ct Scan ◽  
Laboratory Tests ◽  
Anatomic Variation ◽  
Middle Turbinate ◽  
Anatomic Variations ◽  
Concha Bullosa ◽  
Normal Limits ◽  
Osteomeatal Complex ◽  
The Right

ABSTRACT Introduction Concha bullosa is the most common anatomic variation of osteomeatal complex region that is generally seen in the middle turbinate (MT). Materials and methods A 25-year-old male presented with headache and nasal obstruction. Computed tomography (CT) scan documented right paradoxical MT. The right MT also showed aerated concha bullosa with narrow right osteomeatal area. Routine preoperative laboratory tests were within normal limits. Results This case of concha bullosa in paradoxically bent MT was reported, described, and could be safely managed endoscopically. Patient was symptom free up to date without any complication, recurrence, or other pathology. Conclusion Computed tomography may easily identify such uncommon anatomic variations of the osteomeatal region. This directs the surgeon attention to these variations as a cause of headache and osteomeatal area obstruction. How to cite this article El-Anwar MW, Ali AI. Concha Bullosa in Paradoxical Middle Turbinate: A New Variation. Clin Rhinol An Int J 2016;9(3):141-142.

scholarly journals DISTRIBUTION OF ANATOMIC VARIATIONS OF THE NOSE AND PARANASAL SINUSES IN KASHMIRI POPULATION BY COMPUTED TOMOGRAPHY SCAN ANALYSIS.

2021 ◽  
pp. 58-60
Author(s):  
Owais Makhdoomi ◽  
Syed Waseem Abass ◽  
Majid Ul Islam Masoodi
Keyword(s):  
Computed Tomography ◽  
Paranasal Sinuses ◽  
Anatomic Variation ◽  
Disease Process ◽  
Inferior Turbinate ◽  
Anatomic Variations ◽  
Wide Range ◽  
Preoperative Ct ◽  
Kashmiri Population ◽  
Nose And Paranasal Sinuses

Background: Knowledge of anatomy constitutes an integral part of the total management of patients with sinonasal diseases. The aim of this study was to obtain the prevalence of sinonasal anatomic variations in the Kashmiri population and to understand their importance and impact on the disease process, as well as their influence on surgical management and outcome. Materials and Methods: This study is a prospective review of retrospectively performed normal computed tomography (CT) scans of the nose and paranasal sinuses in the adult Kashmiri population at SMHS Hospital. The scans were reviewed by two independent observers. Results: The most common anatomic variation after excluding agger nasi cells were pneumatized Crista Galli, which was seen in 69% of the scans. However, the least common variation seen in this series was Pneumatized inferior turbinate, which was encountered in 1.1 % of the cases. Conclusion: A wide range of regional differences in the prevalence of each anatomic variation exists. Understanding the preoperative CT scan is substantially important because it is the roadmap for the sinus surgeon. Detection of anatomic variations is vital for surgical planning and the prevention of complications.

scholarly journals Methylene Blue versus Coil-Based Computed Tomography-Guided Localization of Lung Nodules

2020 ◽  
Vol 68 (06) ◽  
pp. 540-544 ◽  
Author(s):  
Ze-Dong Zhang ◽  
Hua-Long Wang ◽  
Xian-Yan Liu ◽  
Feng-Fei Xia ◽  
Yu-Fei Fu
Keyword(s):  
Computed Tomography ◽  
Methylene Blue ◽  
Thoracoscopic Surgery ◽  
Wedge Resection ◽  
Lung Nodule ◽  
Lung Nodules ◽  
Complication Rates ◽  
Success Rates ◽  
Technical Success ◽  
Ct Guided

Abstract Background Preoperative computed tomography (CT)-guided localization has been shown to significantly improve lung nodule video-assisted thoracoscopic surgery (VATS)-based wedge resection technical success rates. However, at present, there was insufficient research regarding the optimal approaches to localization of these nodules prior to resection. We aimed to compare the relative clinical efficacy of preoperative CT-guided methylene blue and coil-based lung nodule localization. Methods In total, 91 patients with lung nodules were subjected to either CT-guided methylene blue (n = 34) or coil (n = 57) localization and VATS resection from January 2014 to December 2018. We compared baseline data, localization-associated complication rates, as well as the technical success of localization and resection between these two groups of patients. Results In total, 42 lung nodules in 34 patients underwent methylene blue localization, with associated localization and wedge resection technical success rates of 97.6 and 97.6%, respectively. A total of 71 lung nodules in 57 patients underwent coil localization, with associated localization and wedge resection technical success rates of 94.4 and 97.2%, respectively. There were no significant differences in technical success rates of localization or wedge resection between these groups (p = 0.416 and 1.000, respectively). The coil group sustained a longer duration between localization and VATS relative to the methylene blue group (13.2 vs. 2.5 hours, p = 0.003). Conclusion Both methylene blue and coil localization can be safely and effectively implemented for conducting the diagnostic wedge resection of lung nodules. The coil-based approach is compatible with a longer period of time between localization and VATS procedures.

scholarly journals The Influence of Pre-Procedural Imaging and Cystic Duct Cholangiography on Endoscopic Transpapillary Gallbladder Drainage in Acute Cholecystitis

Diagnostics ◽  
2021 ◽  
Vol 11 (7) ◽  
pp. 1286
Author(s):  
Junya Sato ◽  
Kazunari Nakahara ◽  
Yosuke Michikawa ◽  
Ryo Morita ◽  
Keigo Suetani ◽  
...  
Keyword(s):  
Computed Tomography ◽  
Acute Cholecystitis ◽  
Cystic Duct ◽  
Magnetic Resonance Cholangiopancreatography ◽  
Procedure Time ◽  
Technical Success ◽  
Factors Affecting ◽  
Gallbladder Drainage ◽  
Procedural Evaluation ◽  
The Relationship

Endoscopic transpapillary gallbladder drainage (ETGBD) for acute cholecystitis is challenging. We evaluated the influence of pre-procedural imaging and cystic duct cholangiography on ETGBD. Patients who underwent ETGBD for acute cholecystitis were retrospectively examined. The rate of gallbladder contrast on cholangiography, the accuracy of cystic duct direction and location by computed tomography (CT) and magnetic resonance cholangiopancreatography (MRCP), and the relationship between pre-procedural imaging and the technical success of ETGBD were investigated. A total of 145 patients were enrolled in this study. Gallbladder contrast on cholangiography was observed in 29 patients. The accuracy of cystic duct direction and location (proximal or distal, right or left, and cranial or caudal) by CT were, respectively, 79%, 60%, and 58% by CT and 68%, 55%, and 58% by MRCP. Patients showing gallbladder contrast on cholangiography underwent ETGBD with a significantly shorter procedure time and a lower rate of cystic duct injury. No other factors affecting procedure time, technical success, and cystic duct injury were identified. Pre-procedural evaluation of cystic duct direction and location by CT or MRCP was difficult in patients with acute cholecystitis. Patients who showed gallbladder contrast on cholangiography showed a shorter procedure time and a lower rate of cystic duct injury.

scholarly journals A Prospective Multicenter Study of “Inside Stents” for Biliary Stricture: Multicenter Evolving Inside Stent Registry (MEISteR)

2021 ◽  
Vol 10 (13) ◽  
pp. 2936
Author(s):  
Hirofumi Kogure ◽  
Hironari Kato ◽  
Kazumichi Kawakubo ◽  
Hirotoshi Ishiwatari ◽  
Akio Katanuma ◽  
...  
Keyword(s):  
Multicenter Study ◽  
Stent Placement ◽  
Controlled Trial ◽  
Adverse Event Rate ◽  
Success Rates ◽  
Technical Success ◽  
Biliary Strictures ◽  
Prospective Multicenter Study ◽  
Stent Occlusion ◽  
Benign Biliary Strictures

Background: Endoscopic biliary stent placement is the standard of care for biliary strictures, but stents across the papilla are prone to duodenobiliary reflux, which can cause stent occlusion. Preliminary studies of “inside stents” placed above the papilla showed encouraging outcomes, but prospective data with a large cohort were not reported. Methods: This was a prospective multicenter registry of commercially available inside stents for benign and malignant biliary strictures. Primary endpoint was recurrent biliary obstruction (RBO). Secondary endpoints were technical success of stent placement and removal, adverse events, and stricture resolution. Results: A total of 209 inside stents were placed in 132 (51 benign and 81 malignant) cases with biliary strictures in 10 Japanese centers. During the follow-up period of 8.4 months, RBO was observed in 19% of benign strictures. The RBO rate was 49% in malignant strictures, with the median time to RBO of 4.7 months. Technical success rates of stent placement and removal were both 100%. The adverse event rate was 8%. Conclusion: This prospective multicenter study demonstrated that inside stents above the papilla were feasible in malignant and benign biliary strictures, but a randomized controlled trial is warranted to confirm its superiority to conventional stents across the papilla.

scholarly journals Anatomic variations of internal jugular vein, inferior petrosal sinus and its confluence pattern: Implications in inferior petrosal sinus catheterization

2015 ◽  
Vol 21 (6) ◽  
pp. 769-773 ◽  
Author(s):  
Xianli Lv ◽  
Zhongxue Wu
Keyword(s):  
Computed Tomography ◽  
Internal Jugular Vein ◽  
Jugular Vein ◽  
Treatment Strategies ◽  
Diagnosis And Treatment ◽  
Inferior Petrosal Sinus ◽  
Anatomic Variations ◽  
Review Of The Literature ◽  
Spiral Computed ◽  
Intracranial Lesions

Objective The purpose of this study is to describe anatomic variations of the internal jugular vein (IJV), inferior petrosal sinus (IPS) and their confluence pattern and implications in IPS catheterization. The anatomic route of IPS after going out of the cranium and its confluence patterns with IJV and will supply knowledge about typing of IPS-IJV junction. Method A review of the literature was performed. Results There might be different routes for entering the intracranial segment of the IPS and multislice spiral computed tomography (MSCT) is effective in identifying the confluences of the IPS with the IJV and their courses. It is important to find the confluence of IPS with IJV for diagnosis and treatment of intracranial lesions via venous route. Meanwhile, IPS diameter at the confluence can significantly affect success of catheterization. Conclusion The classification and the theory of the development of the caudal end of the IPS may be useful in establishing treatment strategies that involve endovascular manipulation via the IPS.

scholarly journals Different chest CT scoring systems in patients with COVID-19: could baseline CT be a helpful tool in predicting survival in patients with matched ages and co-morbid conditions?

2021 ◽  
pp. 028418512110063
Author(s):  
Okan Dilek ◽  
Emin Demirel ◽  
Hüseyin Akkaya ◽  
Mehmet Cenk Belibagli ◽  
Gokhan Soker ◽  
...  
Keyword(s):  
Computed Tomography ◽  
Severity Score ◽  
Area Under The Curve ◽  
Scoring Systems ◽  
Survival Prediction ◽  
Success Rates ◽  
Survival Times ◽  
The Mean ◽  
Early Decision ◽  
Age Range

Background Computed tomography (CT) gives an idea about the prognosis in patients with COVID-19 lung infiltration. Purpose To evaluate the success rates of various scoring methods utilized in order to predict survival periods, on the basis of the imaging findings of COVID-19. Another purpose, on the other hand, was to evaluate the agreements among the evaluating radiologists. Material and Methods A total of 100 cases of known COVID-19 pneumonia, of which 50 were deceased and 50 were living, were included in the study. Pre-existing scoring systems, which were the Total Severity Score (TSS), Chest Computed Tomography Severity Score (CT-SS), and Total CT Score, were utilized, together with the Early Decision Severity Score (ED-SS), which was developed by our team, to evaluate the initial lung CT scans of the patients obtained at their initial admission to the hospital. The scans were evaluated retrospectively by two radiologists. Area under the curve (AUC) values were acquired for each scoring system, according to their performances in predicting survival times. Results The mean age of the patients was 61 ± 14.85 years (age range = 18–87 years). There was no difference in co-morbidities between the living and deceased patients. The survival predicted AUC values of ED-SS, CT-SS, TSS, and Total CT Score systems were 0.876, 0.823, 0.753, and 0.744, respectively. Conclusion Algorithms based on lung infiltration patterns of COVID-19 may be utilized for both survival prediction and therapy planning.

Sign in / Sign up

Export Citation Format

Share Document