MyoVision-US: an Artificial Intelligence-Powered Software for Automated Analysis of Skeletal Muscle Ultrasonography.

Introduction/Aims: Muscle ultrasound has high utility in clinical practice and research; however, the main challenges are the training and time required for manual analysis to achieve objective quantification of morphometry. This study aimed to develop and validate a software tool powered by artificial intelligence (AI) by measuring its consistency and predictability of expert manual analysis quantifying lower limb muscle ultrasound images across healthy, acute, and chronic illness subjects. Methods: Quadriceps complex (QC [rectus femoris and vastus intermedius]) and tibialis anterior (TA) muscle ultrasound images of healthy, intensive care unit, and/or lung cancer subjects were captured with portable devices. Automated analyses of muscle morphometry were performed using a custom-built deep-learning model (MyoVision-US), while manual analyses were performed by experts. Consistency between manual and automated analyses was determined using intraclass correlation coefficients (ICC), while predictability of MyoVision -US was calculated using adjusted linear regression (adj.R 2 ). Results: Manual analysis took approximately 24 hours to analyze all 180 images, while MyoVision - US took 247 seconds, saving roughly 99.8%. Consistency between the manual and automated analyses by ICC was good to excellent for all QC (ICC:0.85-0.99) and TA (ICC:0.93-0.99) measurements, even for critically ill (ICC:0.91-0.98) and lung cancer (ICC:0.85-0.99) images. The predictability of MyoVision-US was moderate to strong for QC (adj.R 2 :0.56-0.94) and TA parameters (adj.R 2 :0.81-0.97). Discussion: The application of AI automating lower limb muscle ultrasound analyses showed excellent consistency and strong predictability compared with human analysis. Future work needs to explore AI-powered models for the evaluation of other skeletal muscle groups.

Transcranial ultrasonography to detect intracranial pathology: A systematic review and meta-analysis.

BACKGROUND AND PURPOSE: Transcranial ultrasonography (TCU) can be a useful diagnostic tool in evaluating intracranial pathology in patients with limited or delayed access to routine neuroimaging in critical care or austere settings. We reviewed available literature investigating the diagnostic utility of TCU for detecting pediatric and adult patient's intracranial pathology in patients with intact skulls and reported diagnostic accuracy measures. METHODS: We performed a systematic review of PubMed® , Cochrane Library, Embase® , Scopus® , Web of Science™, and Cumulative Index to Nursing and Allied Health Literature databases to identify articles evaluating ultrasound-based detection of intracranial pathology in comparison to routine imaging using broad Medical Subject Heading sets. Two independent reviewers reviewed the retrieved articles for bias using the Quality Assessment of Diagnostic Accuracy Studies tools and extracted measures of diagnostic accuracy and ultrasound parameters. Data were pooled using meta-analysis implementing a random-effects approach to examine the sensitivity, specificity, and accuracy of ultrasound-based diagnosis. RESULTS: A total of 44 studies out of the 3432 articles screened met the eligibility criteria, totaling 2426 patients (Mean age: 60.1 ± 14.52 years). We found tumors, intracranial hemorrhage (ICH), and neurodegenerative diseases in the eligible studies. Sensitivity, specificity, and accuracy of TCU and their 95% confidence intervals were 0.80 (0.72, 0.89), 0.71 (0.59, 0.82), and 0.76 (0.71, 0.82) for neurodegenerative diseases; 0.88 (0.74, 1.02), 0.81 (0.50, 1.12), and 0.94 (0.92, 0.96) for ICH; and 0.97 (0.92, 1.03), 0.99 (0.96, 1.01), and 0.99 (0.97, 1.01) for intracranial masses. No studies reported ultrasound presets. CONCLUSIONS: TCU has a reasonable sensitivity and specificity for detecting intracranial pathology involving ICH and tumors with clinical applications in remote locations or where standard imaging is unavailable. Future studies should investigate ultrasound parameters to enhance diagnostic accuracy in diagnosing intracranial pathology.

Exploratory study to assess feasibility of intracerebral hemorrhage detection by point of care cranial ultrasound.

BACKGROUND: Limited studies have evaluated the use of ultrasound for detection of intracerebral hemorrhage (ICH) using diagnostic ultrasound Transcranial Doppler machines in adults. The feasibility of ICH detection using Point of care Ultrasound (POCUS) machines has not been explored. We evaluated the feasibility of using cranial POCUS B mode imaging performed using intensive care unit (ICU) POCUS device for ICH detection with a secondary goal of mapping optimal imaging technique and brain topography likely to affect sensitivity and specificity of ICH detection with POCUS. MATERIALS AND METHODS: After obtaining IRB approval, a blinded investigator performed cranial ultrasound (Fujifilm, Sonosite® Xporte, transcranial and abdominal presets) through temporal windows on 11 patients with intracerebral pathology within 72 h of last CT/MRI (computed tomography scan/magnetic resonance imaging) brain after being admitted to a neurocritical care unit in Aug 2020 and Nov 2020-Mar 2021. Images were then compared to patient's CT/MRI to inform topography. Inferential statistics were reported. RESULTS: Mean age was 57 (28-77 years) and 6/11 were female. Six patients were diagnosed with ICH, 3 with ischemic stroke, 1 subarachnoid hemorrhage, and 1 brain tumor. The sensitivity and specificity of point of care diagnosis of ICH compared to CT/MRI brain was 100% and 50%, respectively. Mean time between ultrasound scan and CT/MRI was 13.3 h (21 min-39 h). Falx cerebri, choroid calcification and midbrain-related artifacts were the most reproducible hyperechoic signals. Abdominal preset on high gain yielded less artifact than Transcranial Doppler preset for cranial B mode imaging. False positive ICH diagnosis was attributed to intracerebral tumor and midbrain-related artifact. CONCLUSIONS: Our exploratory analysis yielded preliminary data on use of point of care cranial ultrasound for ICH diagnosis to inform imaging techniques, cranial topography on B mode and sample size estimation for future studies to evaluate sensitivity and specificity of cranial POCUS in adult patients. This pilot study is limited by small sample size and over representation of ICH in the study. Cranial POCUS is feasible using POCUS machines and may have potential as a screening tool if validated in adequately powered studies.

Brain topography on adult ultrasound images: Techniques, interpretation, and image library.

BACKGROUND AND PURPOSE: Many studies have explored the possibility of using cranial ultrasound for discerning intracranial pathologies like tumors, hemorrhagic stroke, or subdural hemorrhage in clinical scenarios where computer tomography may not be accessible or feasible. The visualization of intracranial anatomy on B-mode ultrasound is challenging due to the presence of the skull that limits insonation to a few segments on the temporal bone that are thin enough to allow transcranial transmission of sound. Several artifacts are produced by hyperechoic signals inherent in brain and skull anatomy when images are created using temporal windows. METHODS: While the literature has investigated the accuracy of diagnosis of intracranial pathology with ultrasound, we lack a reference source for images acquired on cranial topography on B-mode ultrasound to illustrate the appearance of normal and abnormal structures of the brain and skull. Two investigators underwent hands-on training in Cranial point-of-care ultrasound (c-POCUS) and acquired multiple images from each patient to obtain the most in-depth images of brain to investigate all visible anatomical structures and pathology within 24 hours of any CT/MRI imaging done. RESULTS: Most reproducible structures visible on c-POCUS included bony parts and parenchymal structures. Transcranial and abdominal presets were equivalent in elucidating anatomical structures. Brain pathology like parenchymal hemorrhage, cerebral edema, and hydrocephalus were also visualized. CONCLUSIONS: We present an illustrated anatomical atlas of cranial ultrasound B-mode images acquired in various pathologies in a critical care environment and compare our findings with published literature by performing a scoping review of literature on the subject.

Prolonged Automated Robotic TCD Monitoring in Acute Severe TBI: Study Design and Rationale.

BACKGROUND: Transcranial Doppler ultrasonography (TCD) is a portable, bedside, noninvasive diagnostic tool used for the real-time assessment of cerebral hemodynamics. Despite the evident utility of TCD and the ability of this technique to function as a stethoscope to the brain, its use has been limited to specialized centers because of the dearth of technical and clinical expertise required to acquire and interpret the cerebrovascular parameters. Additionally, the conventional pragmatic episodic TCD monitoring protocols lack dynamic real-time feedback to guide time-critical clinical interventions. Fortunately, with the recent advent of automated robotic TCD technology in conjunction with the automated software for TCD data processing, we now have the technology to automatically acquire TCD data and obtain clinically relevant information in real-time. By obviating the need for highly trained clinical personnel, this technology shows great promise toward a future of widespread noninvasive monitoring to guide clinical care in patients with acute brain injury. METHODS: Here, we describe a proposal for a prospective observational multicenter clinical trial to evaluate the safety and feasibility of prolonged automated robotic TCD monitoring in patients with severe acute traumatic brain injury (TBI). We will enroll patients with severe non-penetrating TBI with concomitant invasive multimodal monitoring including, intracranial pressure, brain tissue oxygenation, and brain temperature monitoring as part of standard of care in centers with varying degrees of TCD availability and experience. Additionally, we propose to evaluate the correlation of pertinent TCD-based cerebral autoregulation indices such as the critical closing pressure, and the pressure reactivity index with the brain tissue oxygenation values obtained invasively. CONCLUSIONS: The overarching goal of this study is to establish safety and feasibility of prolonged automated TCD monitoring for patients with TBI in the intensive care unit and identify clinically meaningful and pragmatic noninvasive targets for future interventions.

Neurologic Complications in the Postoperative Neurosurgery Patient.

PURPOSE OF REVIEW: This article discusses neurologic complications encountered in the postoperative care of neurosurgical patients that are common or key to recognize in the immediate postoperative period. The major neurosurgical subspecialty procedures (cerebrovascular neurosurgery, neuro-oncology, epilepsy neurosurgery, functional neurosurgery, CSF diversion, endovascular neurosurgery, and spinal surgery) are broadly included under craniotomy procedures, endovascular/vascular procedures, and spinal procedures. This article focuses on the range of complications inherent in these approaches with specific scenarios addressed as applicable. RECENT FINDINGS: The morbidity and mortality related to neurosurgical procedures remains high, necessitating ongoing research and quality improvement efforts in perioperative screening, intraoperative management, surgical approaches, and postoperative care of these patients. Emerging research continues to investigate safer and newer options for routine neurosurgical approaches, such as coiling over clipping for amenable aneurysms, endoscopic techniques for transsphenoidal hypophysectomy, and minimally invasive spinal procedures; postoperative monitoring and care of patients after these procedures continues to be a key component in the continuum of care for improving outcomes. SUMMARY: Postoperative care of patients undergoing major neurosurgical procedures is an integral part of many neurocritical care practices. Neurosurgeons often enlist help from neurologists to assist with evaluation, interpretation, and management of complications in routine inpatient settings. Awareness of the common neurologic complications of various neurosurgical procedures can help guide appropriate clinical monitoring algorithms and quality improvement processes for timely evaluation and management of these patients.

Transcranial Doppler Emboli Monitoring for Infective Endocarditis.

BACKGROUND AND PURPOSE: Ischemic stroke can occur in 20-55% of patients with infective endocarditis (IE) with 75% occurring during the first 2 weeks of treatment. CT or MRI brain can diagnose the sequelae of stroke but transcranial Doppler (TCD) can document active embolization. We undertook a retrospective review of our patient cohort and a systematic review of literature to assess the role of TCD in early diagnosis and management of ischemic stroke in IE. METHODS: Retrospective chart review and literature review. RESULTS: We found 89 patients with stroke caused by IE at our institution from December 2011 to April 2018. TCDs were obtained on 26 patients; 16 were abnormal for cerebrovascular abnormalities. Only 4 patients had 30-minute emboli monitoring performed, of which one revealed emboli. We found 3 studies investigating the role of TCDs in IE that showed promise in its use as a predictive tool in stroke risk stratification. CONCLUSIONS: Presence of embolization in the form of high-intensity transient signals (HITS) detected on TCDs can be used for early diagnosis of IE, assessing efficacy of antibiotic therapy, and stratification of stroke risk in IE. This can aid further research into testing preventative interventions for reducing stroke burden in IE such as earlier valvular surgery or vacuum-assisted vegetation extraction.

Development of a Homemade Spinal Model for Simulation to Teach Ultrasound Guidance for Lumbar Puncture.

BACKGROUND: Spinal procedures such as lumbar punctures (LPs), epidurals, and spinal blocks are essential components to clinical practice but are challenging to teach, learn, or practice on real patients due to patient safety and comfort limiting the number of attempts. Resident physicians traditionally learn these spinal procedural skills through observation of a more senior physician before attempting the procedure. Simulation using models can improve providers' competency without introducing an added risk to patients. A difficulty encountered with access to simulation training for such procedures is the limited availability of simulators. While there are several high-quality, commercially available models that mimic the anatomy of lumbar spine, the cost of these models often limits the access to students and practitioners. The other challenge is access to simulators with versatility that can be used for palpation as well as ultrasound (US)-guided procedures. A simulator that can combine practice of both palpation and US-guided modalities would be efficacious in reducing cost to the teaching institutions. We attempted to overcome the access barrier to spinal models by developing an alternative that provides a good simulator for both palpation and US-guided LP while keeping the cost low. Our model can be easily manufactured by not only clinicians but also medical students. METHODS: A literature review was conducted to assess the available research and information on the production and use of simulators for practicing LPs and other spinal procedures. Publications queried described the production of models and utilizing the information compiled we devised and fabricated a model. RESULTS: A lumbar spine model was developed using computed tomography spine data of an average-sized male patient without lumbar spine pathology. The model was created using medical imaging processing software and printed on 3D printer using nylon plastic. This model was then utilized by residents, advanced practice providers, and medical students for palpation and US-guided LP simulation training. CONCLUSIONS: An inexpensive reusable non-commercial LP simulator can be an effective method for teaching invasive procedures like LPs, especially if it can be used both for palpation and US-guided procedures. The method outlined here can be easily reproduced in a relatively short amount of time. We recognize one limitation in the widespread dissemination of this technique being access to a 3D printer and digital designs for printing. Future studies will be necessary to determine the efficacy of the homemade LP simulator in teaching neurointensivist in training.

Iatrogenic venous thrombosis secondary to supplemental medicine toxicity.

OBJECTIVE: We described a case of cerebral venous sinus thrombosis in a patient taking multiple supplements as part of a naturopathic anti-aging regimen. METHODS: Case report. RESULTS: The patient presented with a thalamic infarct associated with a thrombus in the vein of Galen. He reported no previous history of endocrinopathy and no known hypercoagulability risk factors. He was treated with therapeutic anticoagulation resulting in improvement. Diagnostic workup revealed hyperthyroidism and gonadotrophic deficiency attributed to significant supplement medicine usage. CONCLUSION: This case highlights a potential risk of dietary supplements. Use of these supplements may be a risk factor for idiopathic cerebral venous thrombosis.

Radiological correlate of ocular flutter in a case with paraneoplastic encephalitis.

We present an interesting [18F]fluoro-2-deoxyglucose positron emission tomography (FDG-PET) imaging finding in a patient with ocular flutter and cerebellar ataxia as part of anti-Ma 1/2 antibody-mediated paraneoplastic syndrome associated with a testicular seminoma. He had a typical anterior mesial temporal hyperintensity on magnetic resonance imaging (MRI) without gadolinium enhancement. In addition, his FDG-PET images showed increased deep cerebellar and inferior rectus and superior oblique ocular muscles FDG uptake. This case is the first to visualize in vivo the possible underlying neuropathological mechanism of ocular flutter associated with cerebellar nuclei on functional imaging.