Detection of COVID19 from X-ray images using multiscale Deep Convolutional Neural Network.

The Coronavirus disease 2019 (COVID19) pandemic has led to a dramatic loss of human life worldwide and caused a tremendous challenge to public health. Immediate detection and diagnosis of COVID19 have lifesaving importance for both patients and doctors. The availability of COVID19 tests increased significantly in many countries, thereby provisioning a limited availability of laboratory test kits Additionally, the Reverse Transcription-Polymerase Chain Reaction (RT-PCR) test for the diagnosis of COVID 19 is costly and time-consuming. X-ray imaging is widely used for the diagnosis of COVID19. The detection of COVID19 based on the manual investigation of X-ray images is a tedious process. Therefore, computer-aided diagnosis (CAD) systems are needed for the automated detection of COVID19 disease. This paper proposes a novel approach for the automated detection of COVID19 using chest X-ray images. The Fixed Boundary-based Two-Dimensional Empirical Wavelet Transform (FB2DEWT) is used to extract modes from the X-ray images. In our study, a single X-ray image is decomposed into seven modes. The evaluated modes are used as input to the multiscale deep Convolutional Neural Network (CNN) to classify X-ray images into no-finding, pneumonia, and COVID19 classes. The proposed deep learning model is evaluated using the X-ray images from two different publicly available databases, where database A consists of 1225 images and database B consists of 9000 images. The results show that the proposed approach has obtained a maximum accuracy of 96% and 100% for the multiclass and binary classification schemes using X-ray images from dataset A with 5-fold cross-validation (CV) strategy. For dataset B, the accuracy values of 97.17% and 96.06% are achieved using multiscale deep CNN for multiclass and binary classification schemes with 5-fold CV. The proposed multiscale deep learning model has demonstrated a higher classification performance than the existing approaches for detecting COVID19 using X-ray images.

Augmented Reality in Neurosurgery: A Review of Current Concepts and Emerging Applications.

Augmented reality (AR) superimposes computer-generated virtual objects onto the user's view of the real world. Among medical disciplines, neurosurgery has long been at the forefront of image-guided surgery, and it continues to push the frontiers of AR technology in the operating room. In this systematic review, we explore the history of AR in neurosurgery and examine the literature on current neurosurgical applications of AR. Significant challenges to surgical AR exist, including compounded sources of registration error, impaired depth perception, visual and tactile temporal asynchrony, and operator inattentional blindness. Nevertheless, the ability to accurately display multiple three-dimensional datasets congruently over the area where they are most useful, coupled with future advances in imaging, registration, display technology, and robotic actuation, portend a promising role for AR in the neurosurgical operating room.

The pedicle screw accuracy using a robotic system and measured by a novel three-dimensional method.

Robotics in medicine is associated with precision, accuracy, and replicability. Several robotic systems are used in spine surgery. They are all considered shared-control systems, providing "steady-hand" manipulation instruments. Although numerous studies have testified to the benefits of robotic instrumentations, they must address their true accuracy. Our study used the Mazor system under several situations and compared the spatial accuracy of the pedicle screw (PS) insertion and its planned trajectory. We used two cadaveric specimens with intact spinal structures from C7 to S1. PS planning was performed using the two registration methods (preopCT/C-arm or CT-to-fluoroscopy registration). After planning, the implant spatial orientation was defined based on six anatomic parameters using axial and sagittal CT images. Two surgical open and percutaneous access were used to insert the PS. After that, another CT acquisition was taken. Accuracy was classified into optimal, inaccurate and unacceptable according to the degree of screw deviation from its planning using the same spatial orientation method. Based on the type of spatial deviation, we also classified the PS trajectory into 16 pattern errors. Seven (19%) out of 37 implanted screws were considered unacceptable (deviation distances > 2.0 mm or angulation > 5°), and 14 (38%) were inaccurate (> 0.5 mm and ≤ 2.0 mm or > 2.5° and ≤ 5°). CT-to-fluoroscopy registration was superior to preopCT/C-arm (average deviation in 0.9 mm vs. 1.7 mm, respectively, p < 0.003), and percutaneous was slightly better than open but did not reach significance (1.3 mm vs. 1.7 mm, respectively). Regarding pattern error, the tendency was to have more axial than sagittal shifts. Using a quantitative method to categorize the screw 3D position, only 10.8% of the screws were considered unacceptable. However, with a more rigorous concept of inaccuracy, almost half were non-optimal. We also identified that, unlike some previous results, the O-arm registration delivers more accurate implants than the preopCT/C-arm method.

Fuzzy Logic System Implementation on the Performance Parameters of Health Data Management Frameworks.

The development of wireless sensors and wearable devices has led health care services to the new paramount. The extensive use of sensors, nodes, and devices in health care services generate an enormous amount of health data which is generally unstructured and heterogeneous. Many generous methods and frameworks have been developed for efficient data exchange frameworks, security protocols for data security and privacy. However, very less emphasis has been devoted to structuring and interpreting health data by fuzzy logic systems. The wireless sensors and device performances are affected by the remaining battery/energy, which induces uncertainties, noise, and errors. The classification, noise removal, and accurate interoperation of health data are critical for taking accurate diagnosis and decision making. Fuzzy logic system and algorithms were found to be effective and energy efficient in handling the challenges of raw medical data uncertainties and data management. The integration of fuzzy logic is based on artificial intelligence, neural network, and optimization techniques. The present work entails the review of various works which integrate fuzzy logic systems and algorithms for enhancing the performance of healthcare-related apps and framework in terms of accuracy, precision, training, and testing data capabilities. Future research should concentrate on expanding the adaptability of the reasoning component by incorporating other features into the present cloud architecture and experimenting with various machine learning methodologies.

Intraoperative Error Propagation in 3-Dimensional Spinal Navigation From Nonsegmental Registration: A Prospective Cadaveric and Clinical Study.

STUDY DESIGN: Prospective pre-clinical and clinical cohort study. OBJECTIVES: Current spinal navigation systems rely on a dynamic reference frame (DRF) for image-to-patient registration and tool tracking. Working distant to a DRF may generate inaccuracy. Here we quantitate predictors of navigation error as a function of distance from the registered vertebral level, and from intersegmental mobility due to surgical manipulation and patient respiration. METHODS: Navigation errors from working distant to the registered level, and from surgical manipulation, were quantified in 4 human cadavers. The 3-dimensional (3D) position of a tracked tool tip at 0 to 5 levels from the DRF, and during targeting of pedicle screw tracts, was captured in real-time by an optical navigation system. Respiration-induced vertebral motion was quantified from 10 clinical cases of open posterior instrumentation. The 3D position of a custom spinous-process clamp was tracked over 12 respiratory cycles. RESULTS: An increase in mean 3D navigation error of ≥2 mm was observed at ≥2 levels from the DRF in the cervical and lumbar spine. Mean ± SD displacement due to surgical manipulation was 1.55 ± 1.13 mm in 3D across all levels, ≥2 mm in 17.4%, 19.2%, and 38.5% of levels in the cervical, thoracic, and lumbar spine, respectively. Mean ± SD respiration-induced 3D motion was 1.96 ± 1.32 mm, greatest in the lower thoracic spine (P < .001). Tidal volume and positive end-expiratory pressure correlated positively with increased vertebral displacement. CONCLUSIONS: Vertebral motion is unaccounted for during image-guided surgery when performed at levels distant from the DRF. Navigating instrumentation within 2 levels of the DRF likely minimizes the risk of navigation error.

Optical Topographic Imaging for Spinal Intraoperative Three-Dimensional Navigation in Mini-Open Approaches: A Prospective Cohort Study of Initial Preclinical and Clinical Feasibility.

OBJECTIVE: Computer-assisted three-dimensional navigation often guides spinal instrumentation. Optical topographic imaging (OTI) offers comparable accuracy and significantly faster registration relative to current navigation systems in open posterior thoracolumbar exposures. We validate the usefulness and accuracy of OTI in minimally invasive spinal approaches. METHODS: Mini-open midline posterior exposures were performed in 4 human cadavers. Square exposures of 25, 30, 35, and 40 mm were registered to preoperative computed tomography imaging. Screw tracts were fashioned using a tracked awl and probe with instrumentation placed. Navigation data were compared with screw positions on postoperative computed tomography imaging, and absolute translational and angular deviations were computed. In vivo validation was performed in 8 patients, with mini-open thoracolumbar exposures and percutaneous placement of navigated instrumentation. Navigated instrumentation was performed in the previously described manner. RESULTS: For 37 cadaveric screws, absolute translational errors were (1.79 ± 1.43 mm) and (1.81 ± 1.51 mm) in the axial and sagittal planes, respectively. Absolute angular deviations were (3.81 ± 2.91°) and (3.45 ± 2.82°), respectively (mean ± standard deviation). The number of surface points registered by the navigation system, but not exposure size, correlated positively with the likelihood of successful registration (odds ratio, 1.02; 95% confidence interval, 1.009-1.024; P < 0.001). Fifty-five in vivo thoracolumbar pedicle screws were analyzed. Overall (mean ± standard deviation) axial and sagittal translational errors were (1.79 ± 1.41 mm) and (2.68 ± 2.26 mm), respectively. Axial and sagittal angular errors were (3.63° ± 2.92°) and (4.65° ± 3.36°), respectively. There were no radiographic breaches >2 mm or any neurovascular complications. CONCLUSIONS: OTI is a novel navigation technique previously validated for open posterior exposures and in this study has comparable accuracy for mini-open minimally invasive surgery exposures. The likelihood of successful registration is affected more by the geometry of the exposure than by its size.

Optical Topographic Imaging for Spinal Intraoperative 3-Dimensional Navigation in the Cervical Spine: Initial Preclinical and Clinical Feasibility.

OF BACKGROUND DATA: Computer-assisted 3-dimensional navigation may guide spinal instrumentation. Optical topographic imaging (OTI) is a novel navigation technique offering comparable accuracy and significantly faster registration workflow relative to current navigation systems. It has previously been validated in open posterior thoracolumbar exposures. OBJECTIVE: To validate the utility and accuracy of OTI in the cervical spine. STUDY DESIGN: This is a prospective preclinical cadaveric and clinical cohort study. METHODS: Standard midline open posterior cervical exposures were performed, with segmental OTI registration at each vertebral level. In cadaveric testing, OTI navigation guidance was used to track a drill guide for cannulating screw tracts in the lateral mass at C1, pars at C2, lateral mass at C3-6, and pedicle at C7. In clinical testing, translaminar screws at C2 were also analyzed in addition. Planned navigation trajectories were compared with screw positions on postoperative computed tomographic imaging, and quantitative navigation accuracies, in the form of absolute translational and angular deviations, were computed. RESULTS: In cadaveric testing (mean±SD) axial and sagittal translational navigation errors were (1.66±1.18 mm) and (2.08±2.21 mm), whereas axial and sagittal angular errors were (4.11±3.79 degrees) and (6.96±5.40 degrees), respectively.In clinical validation (mean±SD) axial and sagittal translational errors were (1.92±1.37 mm) and (1.27±0.97 mm), whereas axial and sagittal angular errors were (3.68±2.59 degrees) and (3.47±2.93 degrees), respectively. These results are comparable to those achieved with OTI in open thoracolumbar approaches, as well as using current spinal neuronavigation systems in similar applications. There was no radiographic facet, canal or foraminal violations, nor any neurovascular complications. CONCLUSIONS: OTI is a novel navigation technique allowing efficient initial and repeat registration. Accuracy even in the more mobile cervical spine is comparable to current spinal neuronavigation systems.

High-Frequency Micro-Ultrasound Imaging and Optical Topographic Imaging for Spinal Surgery: Initial Experiences.

High frequency micro-ultrasound (µUS) transducers with central frequencies up to 50 MHz facilitate dynamic visualization of patient anatomy with minimal disruption of the surgical work flow. Micro-ultrasound improves spatial resolution over conventional ultrasound imaging from millimeter to micrometer, but compromises depth penetration. This trade-off is sufficient during an open surgery in which the bone is removed and theultrasound probe can be placed into the surgical cavity. By fusing µUS with pre-operative imaging and tracking the ultrasound probe intra-operatively using our optical topographic imaging technology, we can provide dynamic feedback during surgery, thus affecting clinical decision making. We present our initial experience using high-frequency µUS imaging during spinal procedures. Micro-ultrasound images were obtained in five spinal procedures. Medical rationale for use of µUS was provided for each patient. Surgical procedures were performed using the standard clinical practice with bone removal to facilitate real-time ultrasound imaging of the soft tissue. During surgery, the µUS probe was registered to the pre-operative computed tomography and magnetic resonance images. Images obtained comprised five spinal decompression surgeries (four tumor resections, one cystic synovial mass). Micro-ultrasound images obtained during spine surgery delineated exquisite detailing of the spinal anatomy including white matter and gray matter tracts and nerve roots and allowed accurate assessment of the extent of decompression/tumor resection. In conclusion, tracked µUS enables real-time imaging of the surgical cavity, conferring significant qualitative improvement over conventional ultrasound.

High Speed, High Density Intraoperative 3D Optical Topographical Imaging with Efficient Registration to MRI and CT for Craniospinal Surgical Navigation.

Intraoperative image-guided surgical navigation for craniospinal procedures has significantly improved accuracy by providing an avenue for the surgeon to visualize underlying internal structures corresponding to the exposed surface anatomy. Despite the obvious benefits of surgical navigation, surgeon adoption remains relatively low due to long setup and registration times, steep learning curves, and workflow disruptions. We introduce an experimental navigation system utilizing optical topographical imaging (OTI) to acquire the 3D surface anatomy of the surgical cavity, enabling visualization of internal structures relative to exposed surface anatomy from registered preoperative images. Our OTI approach includes near instantaneous and accurate optical measurement of >250,000 surface points, computed at >52,000 points-per-second for considerably faster patient registration than commercially available benchmark systems without compromising spatial accuracy. Our experience of 171 human craniospinal surgical procedures, demonstrated significant workflow improvement (41 s vs. 258 s and 794 s, p < 0.05) relative to benchmark navigation systems without compromising surgical accuracy. Our advancements provide the cornerstone for widespread adoption of image guidance technologies for faster and safer surgeries without intraoperative CT or MRI scans. This work represents a major workflow improvement for navigated craniospinal procedures with possible extension to other image-guided applications.

Outcome Evaluation of Acute Ischemic Stroke Patients Treated with Endovascular Thrombectomy: A Single-Institution Experience in the Era of Randomized Controlled Trials.

BACKGROUND: Endovascular thrombectomy is an effective procedure to treat selected ischemic strokes, as shown in recent randomized controlled trials (RCTs). The generalizability of these trial data to real-world settings, however, is unknown. The aim of this study was to examine our single-center experience with endovascular thrombectomy for acute ischemic strokes and perform a comparative outcome analysis to the most recent RCTs. METHODS: We performed a 5-year retrospective analysis, from April 2011 to March 2016, on 66 consecutive patients with acute ischemic stroke who received endovascular thrombectomy at our institution. The Alberta Stroke Program Early CT Score (ASPECTS) and the National Institutes of Health Stroke Scale were used to assess preoperative status. Our primary outcomes were the modified Rankin Score (mRS) at 90 days and recanalization grade measured by the 6-point thrombolysis in cerebral infarction (TICI) grading system. RESULTS: Sixty-six patients received endovascular treatment during the study period. Among the patients examined, 35 (53%) had a favorable outcome (mRS 0-2 at 90 days), 23 (35%) a poor outcome (mRS 3-5), and 8 (12%) died. Successful recanalization (TICI score 3-5) was achieved in 68% of cases. In univariate analysis, patients with good outcome at 90 days had significantly greater ASPECTS, lower National Institutes of Health Stroke Scale, and higher TICI scores. In a multiple logistic regression model, higher ASPECTS and TICI scores were significantly and independently associated with favorable outcome. CONCLUSIONS: Excellent outcomes, as demonstrated by the recent RCTs, can be achieved in clinical practice and reproduced in dedicated tertiary centers.

Impact of Smartphone Applications on Timing of Endovascular Therapy for Ischemic Stroke: A Preliminary Study.

BACKGROUND: The metrics of imaging-to-puncture and imaging-to-reperfusion were recently found to be associated with the clinical outcomes of endovascular thrombectomy for acute ischemic stroke. However, measures for improving workflow within hospitals to achieve better timing results are largely unexplored for endovascular therapy. The aim of this study was to examine our experience with a novel smartphone application developed in house to improve our timing metrics for endovascular treatment. METHODS: We developed an encrypted smartphone application connecting all stroke team members to expedite conversations and to provide synchronized real-time updates on the time window from stroke onset to imaging and to puncture. The effects of the application on the timing of endovascular therapy were evaluated with a secondary analysis of our single-center cohort. Our primary outcome was imaging-to-puncture time. We assessed the outcomes with nonparametric tests of statistical significance. RESULTS: Forty-five patients met our criteria for analysis among 66 consecutive patients with acute ischemic stroke who received endovascular therapy at our institution. After the implementation of the smartphone application, imaging-to-puncture time was significantly reduced (preapplication median time, 127 minutes; postapplication time, 69 minutes; P < 0.001). Puncture-to-reperfusion time was not affected by the application use (42 minutes vs. 36 minutes). CONCLUSION: The use of smartphone applications may reduce treatment times for endovascular therapy in acute ischemic stroke. Further studies are needed to confirm our findings.

Spinal intraoperative three-dimensional navigation: correlation between clinical and absolute engineering accuracy.

BACKGROUND CONTEXT: Spinal intraoperative computer-assisted navigation (CAN) may guide pedicle screw placement. Computer-assisted navigation techniques have been reported to reduce pedicle screw breach rates across all spinal levels. However, definitions of screw breach vary widely across studies, if reported at all. The absolute quantitative error of spinal navigation systems is theoretically a more precise and generalizable metric of navigation accuracy. It has also been computed variably and reported in less than a quarter of clinical studies of CAN-guided pedicle screw accuracy. PURPOSE: This study aimed to characterize the correlation between clinical pedicle screw accuracy, based on postoperative imaging, and absolute quantitative navigation accuracy. DESIGN/SETTING: This is a retrospective review of a prospectively collected cohort. PATIENT SAMPLE: We recruited 30 patients undergoing first-time posterior cervical-thoracic-lumbar-sacral instrumented fusion±decompression, guided by intraoperative three-dimensional CAN. OUTCOME MEASURES: Clinical or radiographic screw accuracy (Heary and 2 mm classifications) and absolute quantitative navigation accuracy (translational and angular error in axial and sagittal planes). METHODS: We reviewed a prospectively collected series of 209 pedicle screws placed with CAN guidance. Each screw was graded clinically by multiple independent raters using the Heary and 2 mm classifications. Clinical grades were dichotomized per convention. The absolute accuracy of each screw was quantified by the translational and angular error in each of the axial and sagittal planes. RESULTS: Acceptable screw accuracy was achieved for significantly fewer screws based on 2 mm grade versus Heary grade (92.6% vs. 95.1%, p=.036), particularly in the lumbar spine. Inter-rater agreement was good for the Heary classification and moderate for the 2 mm grade, significantly greater among radiologists than surgeon raters. Mean absolute translational-angular accuracies were 1.75 mm-3.13° and 1.20 mm-3.64° in the axial and sagittal planes, respectively. There was no correlation between clinical and absolute navigation accuracy. CONCLUSIONS: Radiographic classifications of pedicle screw accuracy vary in sensitivity across spinal levels, as well as in inter-rater reliability. Correlation between clinical screw grade and absolute navigation accuracy is poor, as surgeons appear to compensate for navigation registration error. Future studies of navigation accuracy should report absolute translational and angular errors. Clinical screw grades based on postoperative imaging may be more reliable if performed in multiple by radiologist raters.