Sample 1_Augmented Reality Reference in APA format Penza, V., Ortiz, J., Mattos, L. S., Forgione, A., & Momi, E. D XXXXXXXXXXDense soft tissue 3D reconstruction refined with super-pixel segmentation...

I have mentioned everything in Steps document.


Sample 1_Augmented Reality Reference in APA format Penza, V., Ortiz, J., Mattos, L. S., Forgione, A., & Momi, E. D. (2016). Dense soft tissue 3D reconstruction refined with super-pixel segmentation for robotic abdominal surgery. International Journal of Computer Assisted Radiology and Surgery, Volum 3, Issue 2, 197-206. URL of the Reference Level of Journal (Q1, Q2, …Qn) Keywords in this Reference DOI 10.1007/s11548-015-1276-0 Journal Level: Q2 Augmented reality, 3D surface reconstruction, Visualization, Depth estimation, Super pixel Segmentation The Name of the Current Solution (Technique/ Method/ Scheme/ Algorithm/ Model/ Tool/ Framework/ ... etc ) The Goal (Purpose) of this Solution & What is the Problem that need to be solved What are the components of it? Techniques: · Stereoscopic imaging · 3D reconstruction · Non-parametric modified census transformation · Simple Linear Iterative Clustering (SLIC) · (super pixel segmentation) Tools: CT Scan OpenCV Stereoscopic Cameras (288x360) Applied Area: Liver Surgery Problem: the author identified the problem in the complexity of minimally invasive surgery where surgeons are limited to movements and visualization of dense soft tissues accurately. Purpose (Goal): to propose a system that will enhance intra-operative visualization by exploiting 3D information about the surgical site using SLIC refinement. Pre-Operation: · CT Scan · Extraction of organ models · Creation of virtual environment · Real-virtual camera calibration During Surgery: · Real-virtual patient registration · Dense 3D reconstruction The Process (Mechanism) of this Work; Means How the Problem has Solved & Advantage & Disadvantage of Each Step in This Process Process Steps - Pre Operation Advantage Disadvantage (Limitation) 1 CT Scan 2 Extraction of 3D surface models 3 Real-virtual camera calibration Calibrated using OpenCV Library It is able to remove artifacts from the images; thus enhancing quality. Process Steps - During Surgery Advantage Disadvantage (Limitation) 1 Semi-automatic registration of virtual environment on the real patient (not stated if uses machine learning). They use “virtual assistance system” This semi-automatic registration is done using real-virtual camera plane and the view of the abdomen where it is possible to run the visualization. Produces the view of the model for visualization 2 Soft tissue 3D surface reconstruction Composed of 2 different approaches in which the author called it “Method 1” and “Method 2” 3 Equalization Histogram Equalization Used as (equalizedHist) in OpenCV Library. Reconstruction Method 1 (Consists of 3 steps) 1. Matching Computation Cost Absolute intensity Difference (AD) Matching is done to determine the similarity of the intensity of the pixels of the left and right images to detect and eliminate errors. 2. Aggregation Cost Computation Sum of Absolute Difference (SAD) A function to combine these pixels into a more meaningful data set using sum of absolute difference. 3. Disparity Computation Winner takes all strategy computation (WTA) The disparity computation is responsible for validating the matched pixels on the surface by using winner takes all from the values. Method 1 is a surface reconstruction technique which uses equalizations to improve the details within the surface area. Histogram Equalization – improves the contrast and enhances the details that are over or under exposed. AD – To calculate the similarity between two pixels. SAD – the sum of Absolute difference from the computed value in matching cost WTA – used to find the minimum aggregation cost value at each pixel. Helps validate the correct pixels. Problems in disparity refinement can arise if the motion and tissue deformation is not identified. 4 Reconstruction Method 2 (3 main steps without equalization) 1. Matching cost computation Non parametric census transform Matching is done to determine the similarity of pixels of the images in a moving window for comparison. 2. Aggregation cost computation Sum of Hamming Distance (SHD) A function to combine these pixels into a more meaningful data set using sum of Hamming Distance. 3. Disparity Computation The disparity computation is responsible for validating the matched pixels on the surface by using winner takes all from the values. Method 2 is also a surface reconstruction which does not use equalization like method 1, but it uses NCT to track the moving pixels within the surface area. NCT – converts each pixel inside a moving window into a string of bits that represents neighbour pixels for comparison with central pixel and mean pixels within. SHD – compares the bit of strings that represents the pixels and identifies the position to decrease the computational time of the algorithm. Problems in disparity refinement can arise if the motion and tissue deformation is not identified. 5 The SLIC refinement is a Superpixels segmentation process where it fills the holes disparity map and produces better reconstructed images. This method is applied after the initial disparity map is constructed then will be refined through this algorithm to fill the holes and create a more dense disparity map. SLIC provides better matched pixels because the disparity map will have less holes; thus, will provide better output for visualization in the AR. There is a trade-off between accuracy and computation time. SLIC adds about 0.1s in computation time but will increase the accuracy of matched pixels providing a dense disparity map. Major Impact Factors in this Work Dependent Variable Independent Variable Accuracy of error mapping Using 2 heart phantoms(Method 2 with SLIC refinement which is most accurate), it was able to achieve an accuracy in disparity error mapping of : Heart 1 – 1.75mm Heart 2 – 1.79mm Processing time (computational time) 1.29s / frame and 1.30s / frame (using image resolution of 288x360) SLIC refinement adds about 0.1s for both heart phantoms. Depth Perception (pixels matching) Using 2 heart phantoms(Method 2 with SLIC refinement), it was able to match pixels with the percentage of : Heart 1 – 72.6% matched pixels Heart 2 – 66.5% matched pixels Input and Output Feature of This Solution Contribution & The Value of This Work Input - Pre Operation Output - Pre Operation CT Scan Data Preoperative 3D model Input - During Surgery Output - During Surgery 3D model AR view This proposed enhanced system was able to provide accurate results which was able to address, instrument interaction with tissue, breathing, heartbeat using depth estimation methods (real-time 3D reconstruction with Superpixels refinements). The most notable feature in this system is the ability to create a 3D reconstruction of the preoperative model while in the intraoperative considering other factors such movements, and deformations. This enables better visualizations for the surgeon as they are aware of real-time happenings within the surgery as there are always uncertainty that can happen which are not planned. The main contribution of this method is to be able to provide a “virtual assistive system” in which it is able to do 3D reconstructions which overcomes the limitations of the state of the arts limitation. Reconstructing the soft tissues in real-time will provide better visualization for the surgeons as they will see deformations happening and will update based from the previous frame and not the original CT data. Though the main contribution of the paper is within the segmentations refinement method which is SLIC refinements where it fills the holes of the disparity map by matching the pixels to form a better disparity map that will be inserted in the virtual environment. 1. what in the method could have been better? 2. what in the author analyses were missed? 3. was there a technique that could have been used, or a question that could have been asked, that the researchers did not use or ask? Based on the proposed method, the segmentation phase that does the disparity map refinement added on the computational time. There have been modifications within the SLIC algorithm that increases computational speed and accuracy with pixel matching; thus would improve the overall refinement and the accuracy of the system. Surgical tools may be a hindrance to this proposed solution since large occlusions may cause the 3D reconstruction not to identify the soft tissue or deformation accurately since it relies on the previous data and not the preoperative data. Occlusion handling could be addressed in a form of additional technique or integrate a system that would address the tools such as geometric aware systems where in would register the tools in the preoperative phase. 4. were the conclusion justified and How? Analyse This Work By Critical Thinking The Tools That Assessed this Work The conclusions were justified since the main goal of the proposed system is to enhance visualization of the state of the art by proposing 3D reconstruction methods to defeat the static nature of preoperative CT scans. The used a novel SLIC refinement algorithm to enhance the accuracy of the images in terms of disparity maps that will be used in the virtual environment. They were also able to provide suggestions to improve their system, such stating that they would use a GPU based workstation. The proposed system was able to overcome common problems of AR systems in which it relies heavily on pre-operative data. This solution uses dense 3D reconstruction so that it is able to handle soft tissue deformations happening in real-time. They also use SLIC refinement to improve the accuracy which increased the matched pixels of the disparity map outputs by up to 70% compared to the non-refined disparity maps(images). However, there are still limitations to this system. It focuses mainly on reconstructing the soft tissue deformations but given if there are large occlusions(tools) within the area of the reconstructions, it will not be able to accurately reconstruct the tissue which is changing. This can be solved by using geometrically aware systems which seen in other proposed methods within AR. The experiment was carried out using phantoms hearts where there are 2 methods involved that uses 3D reconstructions. Method 1 with SLIC refinement was able to provide 3.11mm and 72.6% matched pixels within 1.21s. Method 2 with SLIC refinement was able to achieve even better results with 1.75mm and 72.6% matched pixels within 1.29s. Sample 2_Networking Reference in APA format Wu, J., Cheng, B., Wang, M., & Chen, J. (2017). Priority-Aware FEC Coding for High-Definition Mobile Video Delivery Using TCP. IEEE Transactions On Mobile Computing, 16(4), 1090-1106. URL of the Reference Level of Journal (Q1, Q2, …Qn) Keywords in this Reference http://ieeexplore.ieee.org.ezproxy.csu.edu.a u/stamp/stamp.jsp?amumber=7498667  Q2 · High definition mobile video  · TCP · Stringent delay constraint · Forward error correction · Priority-awareness · Wireless networks The Name of the Current Solution (Technique/ Method/ Scheme/ Algorithm/ Model/ Tool/ Framework/ ... etc ) The Goal (Purpose) of this Solution & What is the Problem that need to be solved What are the components of it? Techniques: Priority Aware and TCP Oriented coding (PATON) Problem: delivering HD video using TCP over wireless network seems challenging because of high transmission rate and stringent delay constraint of HD video, bandwidth limited and error prone wireless network, and the retransmission scheme of TCP that causes delay. Goal: to