Poh Sun Goh

Medical Image Segmentation Using K-Means Clustering and Improved Watershed Algorithm

We propose a methodology that incorporates k-means and improved watershed segmentation algorithm for medical image segmentation. The use of the conventional watershed algorithm for medical image analysis is widespread because of its advantages, such as always being able to produce a complete division of the image. However, its drawbacks include over-segmentation and sensitivity to false edges. We address the drawbacks of the conventional watershed algorithm when it is applied to medical images by using k-means clustering to produce a primary segmentation of the image before we apply our improved watershed segmentation algorithm to it. The k-means clustering is an unsupervised learning algorithm, while the improved watershed segmentation algorithm makes use of automated thresholding on the gradient magnitude map and post-segmentation merging on the initial partitions to reduce the number of false edges and over-segmentation. By comparing the number of partitions in the segmentation maps of 50 images, we showed that our proposed methodology produced segmentation maps which have 92% fewer partitions than the segmentation maps produced by the conventional watershed algorithm

Review of orbital imaging.

CT and MRI are commonly used in the evaluation of patients with suspected orbital disease. Many different diseases may present within this small anatomical space. The purpose of this article is to present a diagnostic strategy based on a compartment model. Localizing pathology to sinus, bone, extraconal space, muscle cone, intraconal space, optic nerve, globe or lacrimal fossa allows significant reduction in the number of differential diagnoses as these compartments contain different tissues which disease may involve or arise from. Certain diseases may also present in multiple compartments. Common diseases which might present in one or multiple compartments will be discussed.

Tear flow dynamics in the human nasolacrimal ducts--a pilot study using dynamic magnetic resonance imaging.

BackgroundJones’s theory of tear drainage suggests that the lacrimal sac fills when the eyelids are closed and empties into the nasolacrimal duct when the eyelids are open. This is aided by the contraction of the orbicularis muscle during each blink. This study was undertaken to ascertain the possibility of seeing the dynamic movement of tears in the nasolacrimal system during blinks using magnetic resonance dacryocystography (MR-DCG).MethodsThe sac was initially localized with a three-plane gradient echo sequence using a 1.5-T MRI platform. Fast, dynamic MR-DCG was carried out after we had instilled topically balanced salt solution (BSS) in five subjects and 0.5% gadolinium in seven subjects. The volunteers were asked to close and open their eyes during the fast imaging. The images were digitized to enable us to see the actual movement of fluid in the system.ResultsThe tear movement was clearly seen as a bolus in the volunteers where BSS was used. The fluid passed into the nasolacrimal duct after several blinks when patent. The sac was never seen to empty completely. Though differential filling was seen between the upper and lower part of the sac, it was difficult to see the actual fluid movement in the volunteers when topical gadolinium was used.ConclusionThe findings of the study are supportive of the notion of fluid travel in the form of a bolus through the sac. Once a threshold volume is reached in the lower end of the sac, the fluid is seen to pass through the nasolacrimal duct. This happens after several blinks.

Twelve tips for the effective use of videos in medical education

Abstract Videos can promote learning by either complementing classroom activities, or in self-paced online learning modules. Despite the wide availability of online videos in medicine, it can be a challenge for many educators to decide when videos should be used, how to best use videos, and whether to use existing videos or produce their own. We outline 12 tips based on a review of best practices in curriculum design, current research in multimedia learning and our experience in producing and using educational videos. The 12 tips review the advantages of using videos in medical education, present requirements for teachers and students, discuss how to integrate video into a teaching programme, and describe technical requirements when producing one’s own videos. The 12 tips can help medical educators use videos more effectively to promote student engagement and learning.

Designing and Evaluating an Interactive Multimedia Web-Based Simulation for Developing Nurses’ Competencies in Acute Nursing Care: Randomized Controlled Trial

Background Web-based learning is becoming an increasingly important instructional tool in nursing education. Multimedia advancements offer the potential for creating authentic nursing activities for developing nursing competency in clinical practice. Objective This study aims to describe the design, development, and evaluation of an interactive multimedia Web-based simulation for developing nurses’ competencies in acute nursing care. Methods Authentic nursing activities were developed in a Web-based simulation using a variety of instructional strategies including animation video, multimedia instructional material, virtual patients, and online quizzes. A randomized controlled study was conducted on 67 registered nurses who were recruited from the general ward units of an acute care tertiary hospital. Following a baseline evaluation of all participants’ clinical performance in a simulated clinical setting, the experimental group received 3 hours of Web-based simulation and completed a survey to evaluate their perceptions of the program. All participants were re-tested for their clinical performances using a validated tool. Results The clinical performance posttest scores of the experimental group improved significantly (P<.001) from the pretest scores after the Web-based simulation. In addition, compared to the control group, the experimental group had significantly higher clinical performance posttest scores (P<.001) after controlling the pretest scores. The participants from the experimental group were satisfied with their learning experience and gave positive ratings for the quality of the Web-based simulation. Themes emerging from the comments about the most valuable aspects of the Web-based simulation include relevance to practice, instructional strategies, and fostering problem solving. Conclusions Engaging in authentic nursing activities using interactive multimedia Web-based simulation can enhance nurses’ competencies in acute care. Web-based simulations provide a promising educational tool in institutions where large groups of nurses need to be trained in acute nursing care and accessibility to repetitive training is essential for achieving long-term retention of clinical competency.

Medical image segmentation using watershed segmentation with texture-based region merging

The use of the watershed algorithm for image segmentation is widespread because it is able to produce a complete division of the image. However, it is susceptible to over-segmentation and in medical image segmentation, this meant that that we do not have good representations of the anatomy. We address this issue by thresholding the gradient magnitude image and performing post-segmentation merging on the initial segmentation map. The automated thresholding technique is based on the histogram of the gradient magnitude map while the post-segmentation merging is based on the similarity in textural features (namely angular second moment, contrast, entropy and inverse difference moment) belonging to two neighboring partitions. When applied to the segmentation of various facial anatomical structures from magnetic resonance (MR) images, the proposed method achieved an overlap index of 92.6% compared to manual contour tracings. It is able to merge more than 80% of the initial partitions, which indicates that a large amount of over-segmentation has been reduced. Results produced using watershed algorithm with and without the proposed and proposed post-segmentation merging are presented for comparisons.

Should non-expert clinician examiners be used in objective structured assessment of communication skills among final year medical undergraduates?

Background: Adoption of the objective structured clinical examination may be hindered by shortages of clinicians within a specialty. Clinicians from other specialties should be considered as alternative, non-expert examiners. Aims: We assessed the inter-rater agreement between expert and non-expert clinician examiners in an integrated objective structured clinical examination for final year medical undergraduates. Methods: Pairs of expert and non-expert clinician examiners used a rating checklist to assess students in 8 oral communication stations, representing commonly encountered scenarios from medicine, paediatrics, and surgery. These included breaking bad news, managing an angry relative, taking consent for lumbar puncture; and advising a mother on asthma and febrile fits, and an adult on medication use, lifestyle changes and post-suture care of a wound. 439 students participated in the OSCE (206 in 2005, 233 in 2006). Results: There was good to very good agreement (intraclass coefficient: 0.57–0.79) between expert and non-expert clinician examiners, with 5 out of 8 stations having intraclass coefficients ≥0.70. Variation between paired examiners within stations contributed the lowest variance to student scores. Conclusion: These findings support the use of clinicians from other specialties, as ‘non-expert’ examiners, to assess communication skills, using a standardized checklist, thereby reducing the demand on clinicians’ time.

Lacrimal sac volume measurement during eyelid closure and opening.

Background:  Jones’ theory of tear drainage suggests that the lacrimal sac fills when the eyelids are closed and empties when the eyelids are opened. This study was undertaken to see if there is any change in the volume of the lacrimal sac during eyelid closure and opening using images obtained from magnetic resonance dacryocystography using a topical magnetic resonance contrast agent.

3D Segmentation and Quantification of a Masticatory Muscle from MR Data Using Patient-Specific Models and Matching Distributions

A method is proposed for 3D segmentation and quantification of the masseter muscle from magnetic resonance (MR) images, which is often performed in pre-surgical planning and diagnosis. Because of a lack of suitable automatic techniques, a common practice is for clinicians to manually trace out all relevant regions from the image slices which is extremely time-consuming. The proposed method allows significant time savings. In the proposed method, a patient-specific masseter model is built from a test dataset after determining the dominant slices that represent the salient features of the 3D muscle shape from training datasets. Segmentation is carried out only on these slices in the test dataset, with shape-based interpolation then applied to build the patient-specific model, which serves as a coarse segmentation of the masseter. This is first refined by matching the intensity distribution within the masseter volume against the distribution estimated from the segmentations in the dominant slices, and further refined through boundary analysis where the homogeneity of the intensities of the boundary pixels is analyzed and outliers removed. It was observed that the left and right masseter muscles’ volumes in young adults (28.54 and 27.72cm3) are higher than those of older (ethnic group removed) adults (23.16 and 22.13cm3). Evaluation indicates good agreement between the segmentations and manual tracings, with average overlap indexes for the left and right masseters at 86.6% and 87.5% respectively.

Medical Image Segmentation Using Feature-based GVF Snake

We propose a feature-based GVF snake for medical image segmentation here. Feature-based criteria are introduced for the GVF snake to stop its iterations. Without these criteria, the GVF snake might continue its iterations even though it has converged at the targeted object and result in longer computational time. The feature here is the area of the targeted object. Our proposed method comprises of two stages, namely the training stage and the segmentation stage. In the training stage, we acquire prior knowledge on the relative area of the targeted object from training data. In the segmentation stage, the proposed feature-based GVF snake is applied to segment the object from the image after computing the estimated area of the targeted object. In our proposed method, the GVF snake stops its iterations when the area bounded by its propagation is approximately equal to the estimated area and when it undergoes little change over two consecutive iterations. To illustrate the effectiveness of our proposed method, we applied it to the segmentation of the masseter muscle, which is the strongest jaw muscle, from 2-D magnetic resonance (MR) images. Numerical evaluation done indicates good agreement between the computerized and manual segmentations, with mean overlap of 92%.