scholarly journals COMSeg technique for MRI knee cartilage segmentation

2019 ◽  
Vol 10 (2) ◽  
pp. 147-155 ◽  
Author(s):  
Sulaiman Riza ◽  
Djasmir Marlinawati ◽  
Mohamad Amran Mohd Fahmi
Keyword(s):  
Medical Images ◽  
Female Child ◽  
Physical Work ◽  
Imaging Method ◽  
Knee Cartilage ◽  
Mr Images ◽  
Appropriate Treatment ◽  
Thin Structure ◽  
Images Processing ◽  
Selection Of

Segmentation is one of important methods in medical images processing, particularly as it allows images to be analysed. The method used for segmentation depends on the image problem to be resolved. In this research, knee cartilage needs to be segmented to determine the level of the Osteoarthritis (OA) and for further treatment. Knee cartilage is a soft hyline sponge that is located at the end of the femur, tibia and patella bone to release friction during movement. OA is a knee cartilage problem wherein there is a thinning of the cartilage that results in a shift especially happening between femur and tibia bone causing discomfort and pain. Thinning of the knee cartilage is due to many factors such as age, body weight, genetic, accident, sport injury and extreme use such as physical work. OA can occur to a male or female, child or adult. The effects experienced by patients with OA are such as difficulty to walk, limited movement, and pain in the thin cartilage areas. Monitoring of patients' condition needs to be done to help reduce the problem and thereby enable specialists to perform the appropriate treatment. Imaging is a method used today to monitor the condition of patients with OA. Previous studies showed that MRI is a suitable method for monitoring the condition of patients with OA because of its advantages in visualising knee cartilage more clearly than other imaging methods. Thus, for segmenting the knee cartilage which as mentioned before is an important process in medical images processing, the MR images were selected based on many factors. Segmentation in this study was aimed to obtain the cartilage region to diagnose patient OA level. Various segmentation techniques have been developed by researchers in segmenting the knee cartilage region but they have been unable to segment precisely due to the thin structure of the knee cartilage, especially for patients with intermediate and severe OA. COMSeg technique was developed to segment knee cartilage, especially for those experiencing a normal and intermediate OA and try to implement it to severe OA. The development of this new technique takes into account the imaging method used, the images feature obtained so it can be suitable to process knee image and then selection of an appropriate technique to be applied to the selected images as input.

Characteristics of Polycystic Ovarian Syndrome in Sub-fertile Woman Attending to GOPD in ShSMCH

2019 ◽  
Vol 11 (1) ◽  
pp. 17-23
Author(s):  
Jinnat Ara Islam ◽  
Fatema Ashraf ◽  
Eva Rani Nandi
Keyword(s):  
Polycystic Ovarian Syndrome ◽  
Polycystic Ovaries ◽  
Cross Sectional ◽  
Medical College ◽  
Appropriate Treatment ◽  
Pcos Group ◽  
Symptom Complex ◽  
Ovarian Syndrome ◽  
Selection Of ◽  
Fertile Women

Background: Polycystic ovarian syndrome (PCOS) is a condition characterized by menstrual abnormalities (oligo/amenorrhea) and clinical or biochemical features of hyperandrogenism and may manifest at any age. It is a common cause of female subfertility. All the dimensions of PCOS have not been yet completely explored. Methods: It was a cross sectional comparative study carried out at-GOPD of Shaheed Suhrawardy Medical College & Hospital from January, 2016 to December 2016 on 162 subfertile women. Among them 54 were PCOS group and 108 were non PCOS group. PCOS was diagnosed by (Rotterdam criteria 2003) (i) Oligo or anovulation (ii) hyperandrogenism (iii) Polycystic ovaries. Study was done to evaluate and compare the demographic characteristics, clinical, biochemical and ultrasoundgraphic features of sub-fertile women with and without PCOS. Results: A total of 162 sub-fertile women aged 16-36 years. Mean age was 29.5±5.4. There were significant differences between the two groups in terms of (oligo/amenorrhea), hirsutism, WHR and ovarian ultrasound features. There were no significant differences between two groups in correlations between the level of obesity with the incidence of anovulation, hyperandrogenism or with hormonal features. Conclusion: PCOS is one of the important factors causing Infertility. It is an ill-defined symptom complex needed due attention. There is a need to increase awareness regarding. The clinical features of PCOS are heterogenous thus can be investigated accordingly of selection of appropriate treatment modality. J Shaheed Suhrawardy Med Coll, June 2019, Vol.11(1); 17-23

scholarly journals KCB-Net: A 3D Knee Cartilage and Bone Segmentation Network via Sparse Annotation

2021 ◽  
Author(s):  
Yaopeng Peng ◽  
Hao Zheng ◽  
Fahim Zaman ◽  
Lichun Zhang ◽  
Xiaodong Wu ◽  
...  
Keyword(s):  
Medical Image Segmentation ◽  
Small Subset ◽  
Bone Segmentation ◽  
Knee Cartilage ◽  
Mr Images ◽  
3D Images ◽  
Articular Damage ◽  
Primal Dual ◽  
Model Training ◽  
New Framework

<div>Knee cartilage and bone segmentation is critical for physicians to analyze and diagnose articular damage and knee osteoarthritis (OA). Deep learning (DL) methods for medical image segmentation have largely outperformed traditional methods, but they often need large amounts of annotated data for model training, which is very costly and time-consuming for medical experts, especially on 3D images. In this paper, we report a new knee cartilage and bone segmentation framework, KCB-Net, for 3D MR images based on sparse annotation. KCB-Net selects a small subset of slices from 3D images for annotation, and seeks to bridge the performance gap between sparse annotation and full annotation. Specifically, it first identifies a subset of the most effective and representative slices with an unsupervised scheme; it then trains an ensemble model using the annotated slices; next, it self-trains the model using 3D images containing pseudo-labels generated by the ensemble method and improved by a bi-directional hierarchical earth mover’s distance (bi-HEMD) algorithm; finally, it fine-tunes the segmentation results using the primal-dual Internal Point Method (IPM). Experiments on two 3D MR knee joint datasets (the Iowa dataset and iMorphics dataset) show that our new framework outperforms state-of-the-art methods on full annotation, and yields high quality results even for annotation ratios as low as 5%.<br></div>

Optimization of local shape and appearance probabilities for segmentation of knee cartilage in 3-D MR images

2011 ◽  
Vol 115 (12) ◽  
pp. 1710-1720 ◽  
Author(s):  
Soochahn Lee ◽  
Sang Hyun Park ◽  
Hackjoon Shim ◽  
Il Dong Yun ◽  
Sang Uk Lee
Keyword(s):  
Knee Cartilage ◽  
Mr Images ◽  
Local Shape

Open Source Applications for Image Visualization and Processing in Neuroimaging Training

2015 ◽  
pp. 1319-1332
Author(s):  
Juan A. Juanes ◽  
Pablo Ruisoto ◽  
Alberto Prats-Galino ◽  
Andrés Framiñán
Keyword(s):  
Magnetic Resonance ◽  
Medical Imaging ◽  
Open Source ◽  
Medical Images ◽  
Three Dimensional ◽  
Mr Images ◽  
Image Visualization ◽  
3D Slicer ◽  
Volumetric Reconstruction ◽  
Key Features

The aim of this paper is to demonstrate the major role and potential of three of the most powerful open source computerized tools for the advanced processing of medical images, in the study of neuroanatomy. DICOM images were acquired with radiodiagnostic equipment using 1.5 Tesla Magnetic Resonance (MR) images. Images were further processed using the following applications: first, OsiriXTM version 4.0 32 bits for OS; Second, 3D Slicer version 4.3; and finally, MRIcron, version 6. Advanced neuroimaging processing requires two key features: segmentation and three-dimensional or volumetric reconstruction. Examples of identification and reconstruction of some of the most complex neuroimaging elements such vascular ones and tractographies are included in this paper. The three selected applications represent some of the most versatile technologies within the field of medical imaging.

Surgical and Endovascular Management of Patients with Chronic Mesenteric Ischemia: A Single Center Experience

2017 ◽  
Vol 83 (12) ◽  
pp. 1453-1457 ◽  
Author(s):  
Panagiotis Paliogiannis ◽  
Giorgio C. Ginesu ◽  
Alessandro Fancellu ◽  
Aldo Pischedda ◽  
Mario Maiore ◽  
...  
Keyword(s):  
Mesenteric Ischemia ◽  
Celiac Trunk ◽  
Chronic Mesenteric Ischemia ◽  
Therapeutic Approaches ◽  
Single Center Experience ◽  
Appropriate Treatment ◽  
The Mean ◽  
Clinical Records ◽  
Prognostic Data ◽  
Selection Of

Chronic mesenteric ischemia is a rare intestinal disorder, with a potential evolution toward intestinal infraction. The choice of the appropriate treatment is currently the most crucial issue in the management of patients with chronic mesenteric ischemia. We describe our experience with 16 cases, and we discuss the current diagnostic and therapeutic approaches. A retrospective review of the clinical records was performed, and demographic, clinical, therapeutic, and prognostic data were collected. Six patients were females (37%), and the mean age was 62 years. Postprandial pain was present in all the cases, whereas sitophobia and weight loss were detected in 87 per cent of them. Eight patients were treated with open surgery; no perioperative deaths or relevant complications occurred. One patient had a restenosis of the celiac trunk and superior mesenteric artery 10 months after surgery. No deaths or relevant complications occurred in the remaining patients, who underwent an endovascular procedure. One patient presented a restenosis distal to the vascular stent, whereas two patients died due to comorbidities. The low rates of postoperative morbidity, mortality, and restenosis obtained suggest that surgical or endovascular correction of chronic mesenteric ischemia is satisfactory when performed by experienced surgeons, with an adequate selection of the patients.

Modeling Decisions for Artificial Intelligence

2007 ◽  
Vol 11 (1) ◽  
pp. 3-3
Author(s):  
Vicenç Torra ◽  
Yasuo Narukawa ◽  
Sadaaki Miyamoto
Keyword(s):  
Artificial Intelligence ◽  
Fuzzy Clustering ◽  
Heuristic Algorithms ◽  
Autonomous Mobile Robots ◽  
Special Issue ◽  
Simultaneous Application ◽  
The Third ◽  
Appropriate Treatment ◽  
E Mail ◽  
Selection Of

This special issue presents seven papers that are revised and expanded versions of papers presented at the 2nd International Conference on "Modeling Decisions for Artificial Intelligence" (MDAI). This conference, that took place in Tsukuba (Japan) in July 2005, was the second of the series of MDAI conferences that were initiated in 2004 in Barcelona (Catalonia, Spain). In April 2006, the third edition was held in Tarragona (Catalonia, Spain) and the fourth one is planned in Kitakyushu (Japan) in August 2007. These series of conferences were initiated to foster the use of decision related tools as well as information fusion technologies within artificial intelligence applications. In this issue, we present enhanced version of seven papers presented in the conference. The first paper describes a tool that uses fuzzy logic and neural networks for assigning a treatment to rheumatism. The selection of the appropriate treatment follows oriental medicine. The second paper by Wanyama and Far describes a tool for trade-off analysis to be used in those situations related with decision making in which there is no dominant solution. The third paper is devoted to autonomous mobile robots. The authors describe a multi-layered fuzzy control system for the self-localization of the robot. Two papers devoted to fuzzy clustering follow in this issue. First, one that presents a regularization approach with nonlinear membership weights. One of the proposed methods makes not only possible to perform attraction of data to clusters but also repulsion between different clusters. The second paper on clustering proposes the simultaneous application of homogeneity analysis and fuzzy clustering through the consideration of an appropriate objective function that includes two types of memberships. The sixth paper presents a tool for e-mail classification. The tool brings the name of FIS-CRM that stands for Fuzzy Interrelations and Synonymy Conceptual Representation Model. The issue finishes with a paper on meta-heuristic algorithms for a class of container loading problems. To finish this introduction, we would like to thank the referees for their work on the review process as well as to thank Prof. Hirota, Editor-in-Chief of this journal, for providing us with the opportunity to edit this special issue. The help of Kazuki Ohmori and Kenta Uchino from Fuji Technology Press Ltd. is also acknowledged.

Sign in / Sign up

Export Citation Format

Share Document