Search

Background & Objectives: The cerebral small vessel disease (CSVD) manifest as white matter lesion (WML) in magnetic resonance imaging (MRI). The diffusion tensor imaging (DTI) study shows a lower white matter integrity in white matter lesion (WML) and its penumbra region, suggesting cerebral white matter damage. The aim of this study is to establish correlation between the DTI values of white matter with the age related white matter changes (ARWMC) visual rating score and identify predictors of ARWMC score. Methods: 63 brain MRI images of subjects were selected. Relevant clinical information of the subjects which satisfied QRISK2 risk factors were retrieved from the patients’ folder. DTI parameters were obtained via postprocessing at OsiriX DTImap workstation. Results: FA frontal (r = -0.36; p = 0.003) and AD frontal (r = -0.26; p = 0.040) had a substantial and negative correlation with the ARWMC score. There was a significant positive correlation shown between the ARWMC score and RD frontal (r = 0.30; p = 0.018). There were positive significant association between the ARWMC score and age (b(95%CI): 0.106 (0.061, 0.151); p

Kimura disease (KD) is a rare chronic inflammatory disorder of unknown aetiology that primarily affects the head and neck region with lymph node involvement. Young to middle-aged adult Asian males are predominantly affected. The most common presentation is painless subcutaneous swelling in the head and neck region, while proptosis or orbital involvement is very rarely reported. KD shares some features with other inflammatory and neoplastic disorders, including lymphoma; thus, investigations to confirm the diagnosis should not be delayed. Systemic corticosteroids are commonly used to treat KD and show an excellent response; however, the optimal treatment is still uncertain, and KD has a high recurrence rate. We describe the case of a patient with KD who presented with proptosis and post-auricular swelling, which responded well to oral prednisolone treatment.

Inferior sinus venosus arterial septal defect (ASD) is a rare congenital cardiac anomaly. As with other types of ASDs and patent foramen ovale (PFO), this defect results in arteri-venous shunting with the risk of developing paradoxical embolism (PDE) to the systemic circulation from venous emboli. We report a case of a 67-year-old lady presented to emergency department (ED) with massive pulmonary embolism (PE) and recurrent acute limb ischaemia. On CT pulmonary angiography (CTPA), an incidental finding of inferior sinus venosus ASD was discovered that has caused the PDE. Thrombolysis was administered followed by open mechanical thrombectomy. The patient developed massive lower gastrointestinal bleed post thrombolysis then passed away despite embolization. This case report describes the catastrophic effect of PDE to cause disseminated multisystem thromboembolism. It also emphasizes the importance of early detection of a possible right-to-left shunt in patients presented with recurrent acute limb ischaemia. Transthoracic echocardiography has reduced sensitivity for inferior sinus venosus ASD requiring further assessment with transoesophageal echocardiography in selected cases.

We report three cases of femoral fracture, which subsequently developed osteosarcoma within the course of illness. The first patient presented with a spiral fracture treated with an interlocking femoral nail. He presented five months later with a painful knee swelling and was diagnosed with osteosarcoma. He was asymptomatic despite initial radiographs showed osteolysis at the distal metaphysis. The fracture united well and no evidence of marrow or soft tissue contamination. The second patient had a distal femoral fracture and underwent plate stabilization. Osteosarcoma developed at the united fracture site three years later. Both survived six to seven years without evidence of disease following the standard treatment protocol. The third patient had a closed distal third fracture treated with dynamic compression plating. He presented with an osteoblastic lesion in the proximal femur three years later. There was no initial radiological evidence of osteosarcoma one year before the clinical manifestation of the disease.

Pulmonary underdevelopment is a rare congenital disease which manifests as persistent hemithorax opacification at chest radiography. We present three patients with different types of pulmonary underdevelopment, their imaging features and associated anomalies. Case 1 is a premature neonate with persistent respiratory distress. Further imaging confirmed right pulmonary hypoplasia, associated with a patent foramen ovale, patent ductus arteriosus and vertebral anomalies. Case 2 is a 6-year-old child with corrected anorectal malformation, and recurrent pneumonia. Further imaging confirmed left pulmonary aplasia, associated with an aberrant right subclavian artery and vertebral anomaly. Case 3 is a full term neonate who developed excessive drooling of saliva and respiratory distress. Further imaging confirmed right pulmonary agenesis, associated with an atrial septal defect, patent ductus arteriosus and tracheo-oesophageal fistula. Pulmonary underdevelopment is classified into three types: hypoplasia, aplasia and agenesis. The majority of them have associated anomalies. This condition should be considered a differential diagnosis in paediatric patients with an opaque hemithorax on chest radiography.

A neonate with cyanosis at birth was found to have a rare type of tetralogy of Fallot. Echocardiography showed dextrocardia, left aortic arch with constricting and tortuous patent ductus arteriosus (PDA). Computed tomography angiography thorax revealed visceroatrial situs inversus, mirror image dextrocardia, tetralogy anatomy and tortuous right PDA arising from bovine brachiocephalic artery. In view of severe cyanosis, emergency division of PDA and modified right Blalock-Taussig shunt through median sternotomy were performed under cardiopulmonary bypass. Post-operatively, the sternum was left open and he developed pulmonary overcirculation requiring prolonged ICU stay. He was discharged well at Day 26 post surgery. This case highlights a rare association tetralogy of Fallot, dextrocardia and situs inversus, with concomitant unilateral right PDA and bovine aortic arch.

Manual segmentation is sensitive to operator bias, while semiautomatic random walks segmentation offers an intuitive approach to understand the user knowledge at the expense of large amount of user input. In this paper, we propose a novel random walks seed auto-generation (SAGE) hybrid model that is robust to interobserver error and intensive user intervention. Knee image is first oversegmented to produce homogeneous superpixels. Then, a ranking model is developed to rank the superpixels according to their affinities to standard priors, wherein background superpixels would have lower ranking values. Finally, seed labels are generated on the background superpixel using Fuzzy C-Means method. SAGE has achieved better interobserver DSCs of 0.94 ± 0.029 and 0.93 ± 0.035 in healthy and OA knee segmentation, respectively. Good segmentation performance has been reported in femoral (Healthy: 0.94 ± 0.036 and OA: 0.93 ± 0.034), tibial (Healthy: 0.91 ± 0.079 and OA: 0.88 ± 0.095) and patellar (Healthy: 0.88 ± 0.10 and OA: 0.84 ± 0.094) cartilage segmentation. Besides, SAGE has demonstrated greater mean readers’ time of 80 ± 19 s and 80 ± 27 s in healthy and OA knee segmentation, respectively. SAGE enhances the efficiency of segmentation process and attains satisfactory segmentation performance compared to manual and random walks segmentation. Future works should validate SAGE on progressive image data cohort using OA biomarkers.

BACKGROUND Existing knee cartilage segmentation methods have reported several technical drawbacks. In essence, graph cuts remains highly susceptible to image noise despite extended research interest; active shape model is often constraint by the selection of training data while shortest path have demonstrated shortcut problem in the presence of weak boundary, which is a common problem in medical images. OBJECTIVES The aims of this study is to investigate the capability of random walks as knee cartilage segmentation method. METHODS Experts would scribble on knee cartilage image to initialize random walks segmentation. Then, reproducibility of the method is assessed against manual segmentation by using Dice Similarity Index. The evaluation consists of normal cartilage and diseased cartilage sections which is divided into whole and single cartilage categories. RESULTS A total of 15 normal images and 10 osteoarthritic images were included. The results showed that random walks method has demonstrated high reproducibility in both normal cartilage (observer 1: 0.83±0.028 and observer 2: 0.82±0.026) and osteoarthritic cartilage (observer 1: 0.80±0.069 and observer 2: 0.83±0.029). Besides, results from both experts were found to be consistent with each other, suggesting the inter-observer variation is insignificant (Normal: P=0.21; Diseased: P=0.15). CONCLUSION The proposed segmentation model has overcame technical problems reported by existing semi-automated techniques and demonstrated highly reproducible and consistent results against manual segmentation method.

Pleomorphic adenoma provides as much as 40-70% of tumors in the minor salivary gland with the palate being the most frequent area engaged. Head and neck tumors in parapharyngeal space amount to less than 1%. We share a case of pleomorphic adenoma primarily arising in parapharyngeal space. This report highlights clinical features, pathology, radiological findings, and treatment of this tumor.

Manual segmentation is the standard procedure in osteoarthritis study. However, this method is infamous for being excessive, time consuming and exhaustive. In this study, we overcame the problem of excessive expert interaction reported in manual segmentation by developing a semi-automated random walks technique with computer-aided labelling system. To minimize expert interaction, non-cartilage seeds were generated by using computer. Then, random walks algorithm would segment knee cartilage based on cartilage seeds and non-cartilage seeds. Finally, segmentation results were revised and refined accordingly. A total of 15 normal images and 10 osteoarthritic images were used in this study. In term of efficiency, we have reduced the processing time to segment normal cartilage by 47.5% (93±21s; P = 0.0000019) for observer 1 and 44% (61±8s; P = 3.52±10-5) for observer 2. We also reduced the processing time to segment diseased cartilage by 48.1% (56±16s; P = 0.00014) for observer 1 and 30.3% (62±14s; P = 0.0070) for observer 2. Besides, the proposed technique have produced good reproducibility in both normal (0.83±0.028 for observer 1 and 0.80±0.040 for observer 2) and diseased (0.80±0.060 for observer 1 and 0.82±0.043 for observer 2) cartilage segmentations. In conclusion, the combination of computer generated seeds and user-friendly random walks method have reduced the amount of expert interaction to necessary level without compromising the accuracy of results.

We report a case of a 46-year-old woman who has presented to a peripheral hospital with progressive exertional dyspnoea and chest discomfort. The resting ECG showed features of left-sided ventricular hypertrophy. The initial chest radiograph was reported as cardiomegaly. Initial echocardiography revealed left atrial dilatation and ‘left ventricular’ hypertrophy with normal ejection fraction. She was treated as possible coronary artery disease and was subsequently referred to our centre for CT coronary angiography. Findings from the CT scan were consistent with congenitally corrected transposition of the great arteries (ccTGA). This report describes the radiological features of ccTGA, its associated cardiovascular anomalies, pathophysiology and potential complications.

Segmentation is an instrumental task in medical image analysis. In addition to existing manual, semi-automatic and automatic segmentation models, deep learning has been the niftiest machine learning technique in current research interests. However, none of the models or technique can escape from the overdependence on training data and user intervention. As a result, the use of computer-aided and learning algorithms have reported lackluster robustness in the presence of high anatomical disparity. In recognition of this, we have proposed a binary seeds auto-generation model to reduce the reliance on manually crafted priori information in deep learning. Then, we computed the reproducibility of the proposed model against manual segmentation using normal and osteoarthritic knee magnetic resonance image. In normal knee image, mean agreements of the proposed model and manual segmentation were $0.94 \pm 0.022$ and $0.83 \pm 0.028$ respectively. In osteoarthritic knee image, mean agreements of the proposed model and manual segmentation were $\mathbf{0.92}\pm\mathbf{0.051}$ and $\mathbf{0.79}\pm \mathbf{0.073}$ respectively. Pair t test showed that our method has better accuracy than manual segmentation in both cases (normal: $\text{P}=1.03\times 10^{-9}$; osteoarthritic: $\text{P}=4.94\times 10^{-8}$). Therefore, we can conclude the model is robust to be implemented as part of deep learning based segmentation framework.

Total anomalous pulmonary venous return (TAPVR) is a rare congenital heart defect, and patients are usually symptomatic at a very young age. Survival to adulthood without surgical correction is extremely rare. We report a 33-year-old woman with a heart murmur and a history of a successful pregnancy. Echocardiogram revealed a large atrial septal defect with suspicious pulmonary vein anomaly. Chest radiograph demonstrated classical 'snowman' configuration. Cardiac catheterisation was consistent with anomalous pulmonary venous drainage. Cardiac CT confirmed supracardiac TAPVR, whereby all the pulmonary veins drain into the anomalous vein and finally to the superior vena cava. She remained asymptomatic and underwent a successful surgical repair.

Manual segmentation has been widely accepted in medical image analysis because the user will assume absolute control over final output. Nonetheless, the method is infamous for being time consuming and sensitive to inter-observer variability. Semi-automated methods incorporate user knowledge to initialize computerized segmentation and thus, is a versatile approach to tackle real time segmentation problem. Interactive tool or Graphic User Interface is the most critical medium of communication between user and computer but this element has always been neglected. Intuitively, the design of efficient interactive tool will impose significant effect on segmentation efficiency. In this work, we have presented a mini review on the design element of graphic user interface in medical image segmentation to highlight the importance of this tool in ensuring desirable segmentation output.

Excellent visual quality of medical image is important for subsequent image interpretation and diagnosis planning. Hence, preservation of brightness and considerable tissue contrast improvement are essential for this purpose. Previous histogram equalization methods tend to focus on brightness preservation and neglect the need to address excessive contrast enhancement. However, gray level medical images are highly sensitive to redundant adjustment of image contrast. In this work, we propose to implement cubic spline to reduce the difference in pixel intensity prior to histogram equalization while retaining the partition of histogram to prevent dramatic mean brightness shift in resultant image. Qualitative results indicate that our proposed contrast enhancement method is capable of avoiding extreme brightness elevation and excessive tissue contrast improvement to provide holistic views of the knee MR images. In future, we would like to recommend the development of adaptive data interval to substitute the uniform data point intervals.

In medical image segmentation, manual segmentation is considered both labor- and time-intensive while automated segmentation often fails to segment anatomically intricate structure accordingly. Interactive segmentation can tackle shortcomings reported by previous segmentation approaches through user intervention. To better reflect user intention, development of suitable editing functions is critical. In this paper, we propose an interactive knee cartilage extraction software that covers three important features: intuitiveness, speed, and convenience. The segmentation is performed using multi-label random walks algorithm. Our segmentation software is simple to use, intuitive to normal and osteoarthritic image segmentation and efficient using only two third of manual segmentation's time. Future works will extend this software to three dimensional segmentation and quantitative analysis.

Semi-automatic segmentation is common in medical image processing because anatomical geometries demonstrated by human anatomical parts often requires manual supervision to provide desirable results. However, semi-automatic segmentation has been infamous for requiring excessive human intervention and time consuming. In order to reduce a forementioned problems, seed labels have been generated automatically using superpixels in our previous works. A fixed threshold method has been implemented to classify cartilage and background superpixels but this method is reported to lack the adaptiveness to changing image properties in 3D magnetic resonance image of knee. As a result, the coverage of background seeds are not sufficient to cover whole background area in some cases. In this work, we proposed a local mean based adaptive threshold method as a better alternative to the fixed threshold method. We calculated local mean for each block in an integral image and then use it to differentiate background superpixels from cartilage superpixels. The method is robust to illumination changes and simple to use. We tested the adaptive threshold on 35 knee images of different anatomical geometries and proved the proposed method could provide more comprehensive background seed labels distribution compared to fixed threshold method

Difficulties in analyzing diseased knee cartilage is attributed to large anatomical variation caused by osteoarthritic features. Using expert initialization as crucial source of information, interactive method is very useful in osteoarthritic cartilage segmentation. However, high level of expert intervention remains an issue for existing interactive segmentation model. Therefore, a flexible seeds labelling method has been proposed to automatically cover non-cartilage tissue. Compared to manual segmentation, the proposed method lowers the processing time by 48% in observer 1 and 30% in observer 2. Besides, the proposed method has demonstrated high Dice's reproducibility of 0.80 for observer 1 and 0.82 for observer 2. In future, inter-observer reproducibility should be performed to analyze the progression of knee osteoarthritis.

Unlike automated segmentation, the accuracy of semi-automated segmentation is affected by pertinent parameters such as observer, type of methods and type of cartilage. In this paper, we investigated the effect of these parameters on segmentation results. Based on Dice similarity index obtained from fifteen normal and ten diseased magnetic resonance images, a parameter estimation model was constructed to study the impact of each parameter. Then, we conducted deviance test to verify the effect’s significance. Our result showed that implementation of the proposed segmentation model would introduce positive effect (+0.12) on reproducibility compared to conventional random walks model. Furthermore, we have found intriguing results indicating cartilage normality has diminished effect on reproducibility and tibial cartilage’s result could be influenced by external factors as well. Lastly, our findings highlighted on the necessity of refinement for semi-automated segmentation.

Although interactive segmentation helps lower the degree of human intervention, further upgrade to increase the efficiency and intuitiveness of interactive segmentation method remains requisite. Invariably, newly acquired medical images are not segmented instantly by radiologists due to heavy work load. Therefore, improvement can be made by capitalizing on the time interval between image acquisition and image segmentation. We proposed to pre-generate non-cartilage seeds during this time interval so the labeling process can be simplified. This enhanced interactive segmentation decreases processing time required by a radiologist without compromising the quality of original segmentation quality. Future works should focus on developing automatic pre-generation of different types of cartilage labels.

Knee osteoarthritis is the second most dreadful disease after cardiovascular diseases. Affected patients will not have any effective cure and face the risk of undergoing total knee replacement in chronic stage. Quantitative analysis enhances our understanding of the pathophysiology of osteoarthritis. Nonetheless, manual segmentation is notorious for time- and resource-intensive. Hence, we propose a multilabel, semiautomated segmentation method based on random walks to facilitate the segmentation process. Random walks method is robust to noise, allows multiple objects segmentation and achieves global minimum solution. Our experiment results indicated that random walks achieved greater efficiency than manual segmentation while preserved the quality of knee cartilage segmentation as measured by the Dice's coefficient.

Well-defined image can assist user to identify region of interest during segmentation. However, complex medical image is usually characterized by poor tissue contrast and low background luminance. The contrast improvement can lift image visual quality, but the fundamental contrast enhancement methods often overlook the sudden jump problem. In this work, the proposed bihistogram Bezier curve contrast enhancement introduces the concept of “adequate contrast enhancement” to overcome sudden jump problem in knee magnetic resonance image. Since every image produces its own intensity distribution, the adequate contrast enhancement checks on the image’s maximum intensity distortion and uses intensity discrepancy reduction to generate Bezier transform curve. The proposed method improves tissue contrast and preserves pertinent knee features without compromising natural image appearance. Besides, statistical results from Fisher’s Least Significant Difference test and the Duncan test have consistently indicated that the proposed method outperforms fundamental contrast enhancement methods to exalt image visual quality. As the study is limited to relatively small image database, future works will include a larger dataset with osteoarthritic images to assess the clinical effectiveness of the proposed method to facilitate the image inspection.

Breast cancer is the most common malignancy in female, accounting for nearly 22.9% of all malignancy in female worldwide. However, bilateral breast cancer is rare and sometimes missed. We reported a 44 year old lady presented with late presentation of bilateral invasive lobular breast carcinoma, focusing on her risk factors and treatment strategy.