An End-to-End Platform for Digital Pathology Using Hyperspectral Autofluorescence Microscopy and Deep Learning-Based Virtual Histology.

Conventional histopathology involves expensive and labor-intensive processes that often consume tissue samples, rendering them unavailable for other analyses. We present a novel end-to-end workflow for pathology powered by hyperspectral microscopy and deep learning. First, we developed a custom hyperspectral microscope to nondestructively image the autofluorescence of unstained tissue sections. We then trained a deep learning model to use autofluorescence to generate virtual histologic stains, which avoids the cost and variability of chemical staining procedures and conserves tissue samples. We showed that the virtual images reproduce the histologic features present in the real-stained images using a randomized nonalcoholic steatohepatitis (NASH) scoring comparison study, where both real and virtual stains are scored by pathologists (D.T., A.D.B., R.K.P.). The test showed moderate-to-good concordance between pathologists' scoring on corresponding real and virtual stains. Finally, we developed deep learning-based models for automated NASH Clinical Research Network score prediction. We showed that the end-to-end automated pathology platform is comparable with an independent panel of pathologists for NASH Clinical Research Network scoring when evaluated against the expert pathologist consensus scores. This study provides proof of concept for this virtual staining strategy, which could improve cost, efficiency, and reliability in pathology and enable novel approaches to spatial biology research.

La Crosse Virus Neuroinvasive Disease in Children: A Contemporary Analysis of Clinical/Neurobehavioral Outcomes and Predictors of Disease Severity.

BACKGROUND: La Crosse virus (LACV) is the most common neuroinvasive arboviral infection in children in the United States. However, data regarding predictors of disease severity and neurologic outcome are limited. Additionally, long-term neurologic and neurobehavioral outcomes remain relatively sparse. METHODS: This was a single-center, retrospective cohort study, followed by recruitment for a cross-sectional analysis of long-term neurobehavioral outcomes, among children aged 0-18 years with proven or probable LACV neuroinvasive disease (LACV-ND) between January 2009 and December 2018. Case ascertainment was assured by International Classification of Diseases, Ninth and Tenth Revision, Clinical Modification codes cross-referenced with laboratory results detecting LACV. Demographics, diagnostics, radiographs, and outcomes were evaluated. Recruitment of patients with prior diagnosis of LACV-ND occurred from January 2020 to March 2020, with assessment performed by validated pediatric questionnaires. RESULTS: One-hundred fifty-two children (83 males; median age, 8 years [interquartile range, 5-11.5 years]) were diagnosed with proven (n = 61 [47%]) and probable (n = 91 [60%]) LACV-ND. Sixty-five patients (43%) had severe disease. Altered mental status (AMS) (odds ratio [OR], 6.36 [95% confidence interval {CI}, 2.03-19.95]; P = .0002) and seizures at presentation (OR, 10.31 [95% CI, 3.45-30.86]; P = .0001) were independent predictors of severe disease. Epileptiform discharges on electroencephalogram (EEG) were independently associated with epilepsy diagnosis at follow-up (OR, 13.45 [95% CI, 1.4-128.77]; P = .024). Fifty-four patients were recruited for long-term neurobehavioral follow-up, with frequent abnormal assessments identified (19%-54%) irrespective of disease severity. CONCLUSIONS: Severe disease was observed frequently among children with LACV-ND. Seizures and AMS at presentation were independent predictors of severe disease. EEG may help determine long-term epilepsy risk. Long-term neurobehavioral issues are frequent and likely underrecognized among children with LACV-ND.

Publisher Correction: DeepSTORM3D: dense 3D localization microscopy and PSF design by deep learning.

An amendment to this paper has been published and can be accessed via a link at the top of the paper.

DeepSTORM3D: dense 3D localization microscopy and PSF design by deep learning.

An outstanding challenge in single-molecule localization microscopy is the accurate and precise localization of individual point emitters in three dimensions in densely labeled samples. One established approach for three-dimensional single-molecule localization is point-spread-function (PSF) engineering, in which the PSF is engineered to vary distinctively with emitter depth using additional optical elements. However, images of dense emitters, which are desirable for improving temporal resolution, pose a challenge for algorithmic localization of engineered PSFs, due to lateral overlap of the emitter PSFs. Here we train a neural network to localize multiple emitters with densely overlapping Tetrapod PSFs over a large axial range. We then use the network to design the optimal PSF for the multi-emitter case. We demonstrate our approach experimentally with super-resolution reconstructions of mitochondria and volumetric imaging of fluorescently labeled telomeres in cells. Our approach, DeepSTORM3D, enables the study of biological processes in whole cells at timescales that are rarely explored in localization microscopy.

Urgent procedures or surgeries in patients receiving oral anticoagulants: a systematic literature review.

There are limited data about the frequency of urgent surgical emergencies among patients receiving oral anticoagulants (OACs). We conducted a systematic literature review of Medline and EMBASE for published English-language articles of adult patients receiving oral anticoagulant treatment (vitamin K antagonists, apixaban, dabigatran, edoxaban, rivaroxaban) that reported on patients experiencing unplanned emergent or urgent surgery/procedure or trauma. Randomized trials, observational studies, and case series (50-100 cases) were included. The primary outcome was the frequency of unplanned urgent surgery or invasive procedures among OAC-treated patients with a focus on those not precipitated by the presence of major bleeding. The protocol was not registered. Funding was provided by Covis Pharmaceuticals. The search yielded 1367 potential studies of which 34 were included in the final review. One study reported the rate of urgent surgery/procedures among a large cohort of patients treated with dabigatran or warfarin for atrial fibrillation (~ 1% per year). Another study reported the rate of bleeding or urgent surgery among OAC-treated patients experiencing a fracture or trauma (0.489% per patient-year). The remaining 32 studies were cohorts of OAC-treated patients who received reversal or hemostatic therapies for major bleeding or urgent surgery. A median of 28.8% of these patients underwent surgery or invasive procedure. Urgent surgery appears to be a common, yet understudied complication during OAC treatment potentially associated with high rates of adverse outcomes. With increased eligibility for OACs, future studies evaluating the management and outcomes in this setting are needed.

Ciraparantag reverses the anticoagulant activity of apixaban and rivaroxaban in healthy elderly subjects.

AIMS: Ciraparantag is a reversal agent for anticoagulants including direct oral anticoagulants. The aim was to evaluate the efficacy and safety of ciraparantag to reverse anticoagulation induced by apixaban or rivaroxaban in healthy elderly adults. METHODS AND RESULTS: Two randomized, placebo-controlled, dose-ranging trials conducted in healthy subjects aged 50-75 years. Subjects received apixaban (Study 1) 10 mg orally twice daily for 3.5 days or rivaroxaban (Study 2) 20 mg orally once daily for 3 days. At steady-state anticoagulation subjects were randomized 3:1 to a single intravenous dose of ciraparantag (Study 1: 30, 60, or 120 mg; Study 2: 30, 60, 120, or 180 mg) or placebo. Efficacy was based on correction of the whole blood clotting time (WBCT) at multiple timepoints over 24 h. Subjects and technicians performing WBCT testing were blinded to treatment. Complete reversal of WBCT within 1 h post-dose and sustained through 5 h (apixaban) or 6 h (rivaroxaban) was dose related and observed with apixaban in 67%, 100%, 100%, and 17% of subjects receiving ciraparantag 30 mg, 60 mg, 120 mg, or placebo, respectively; and with rivaroxaban in 58%, 75%, 67%, 100%, and 13% of subjects receiving ciraparantag 30 mg, 60 mg, 120 mg, 180 mg, or placebo, respectively. Adverse events related to ciraparantag were mild, transient hot flashes or flushing. CONCLUSIONS: Ciraparantag provides a dose-related reversal of anticoagulation induced by steady-state dosing of apixaban or rivaroxaban. Sustained reversal was achieved with 60 mg ciraparantag for apixaban and 180 mg ciraparantag for rivaroxaban. All doses of ciraparantag were well tolerated.

Status Epilepticus.

Status epilepticus is a neurologic emergency defined as a seizure that lasts longer than 5 minutes. This is the most common neurologic emergency in children, and it is associated with significant morbidity and mortality. Initial seizure management focuses on stabilization of the patient, followed by medication to terminate the seizure. Benzodiazepines, levetiracetam, fosphenytoin, valproic acid and other antiseizure medications can effectively halt status epilepticus. There is a narrow but important differential diagnosis, including prolonged psychogenic nonepileptic seizure, status dystonicus, and nonconvulsive status epilepticus. Focused laboratory testing, neuroimaging, and electroencephalography can be useful in the evaluation of status epilepticus. Sequelae include focal neurologic deficits, cognitive impairment, and behavioral problems. Pediatricians play an important role in the early recognition and treatment of status epilepticus, thereby preventing the acute and chronic harm that can be associated with status epilepticus.

Artificial intelligence for phase recognition in complex laparoscopic cholecystectomy.

BACKGROUND: The potential role and benefits of AI in surgery has yet to be determined. This study is a first step in developing an AI system for minimizing adverse events and improving patient's safety. We developed an Artificial Intelligence (AI) algorithm and evaluated its performance in recognizing surgical phases of laparoscopic cholecystectomy (LC) videos spanning a range of complexities. METHODS: A set of 371 LC videos with various complexity levels and containing adverse events was collected from five hospitals. Two expert surgeons segmented each video into 10 phases including Calot's triangle dissection and clipping and cutting. For each video, adverse events were also annotated when present (major bleeding; gallbladder perforation; major bile leakage; and incidental finding) and complexity level (on a scale of 1-5) was also recorded. The dataset was then split in an 80:20 ratio (294 and 77 videos), stratified by complexity, hospital, and adverse events to train and test the AI model, respectively. The AI-surgeon agreement was then compared to the agreement between surgeons. RESULTS: The mean accuracy of the AI model for surgical phase recognition was 89% [95% CI 87.1%, 90.6%], comparable to the mean inter-annotator agreement of 90% [95% CI 89.4%, 90.5%]. The model's accuracy was inversely associated with procedure complexity, decreasing from 92% (complexity level 1) to 88% (complexity level 3) to 81% (complexity level 5). CONCLUSION: The AI model successfully identified surgical phases in both simple and complex LC procedures. Further validation and system training is warranted to evaluate its potential applications such as to increase patient safety during surgery.

Detection of elusive polyps using a large-scale artificial intelligence system (with videos).

BACKGROUND AND AIMS: Colorectal cancer is a leading cause of death. Colonoscopy is the criterion standard for detection and removal of precancerous lesions and has been shown to reduce mortality. The polyp miss rate during colonoscopies is 22% to 28%. DEEP DEtection of Elusive Polyps (DEEP2) is a new polyp detection system based on deep learning that alerts the operator in real time to the presence and location of polyps. The primary outcome was the performance of DEEP2 on the detection of elusive polyps. METHODS: The DEEP2 system was trained on 3611 hours of colonoscopy videos derived from 2 sources and was validated on a set comprising 1393 hours from a third unrelated source. Ground truth labeling was provided by offline gastroenterologist annotators who were able to watch the video in slow motion and pause and rewind as required. To assess applicability, stability, and user experience and to obtain some preliminary data on performance in a real-life scenario, a preliminary prospective clinical validation study was performed comprising 100 procedures. RESULTS: DEEP2 achieved a sensitivity of 97.1% at 4.6 false alarms per video for all polyps and of 88.5% and 84.9% for polyps in the field of view for less than 5 and 2 seconds, respectively. DEEP2 was able to detect polyps not seen by live real-time endoscopists or offline annotators in an average of .22 polyps per sequence. In the clinical validation study, the system detected an average of .89 additional polyps per procedure. No adverse events occurred. CONCLUSIONS: DEEP2 has a high sensitivity for polyp detection and was effective in increasing the detection of polyps both in colonoscopy videos and in real procedures with a low number of false alarms. (Clinical trial registration number: NCT04693078.).

Can we predict Acute Medical readmissions using the BOOST tool? A retrospective case note review.

BACKGROUND: Readmissions within 30-days of hospital discharge are a problem. The aim was to determine if the Better Outcomes for Older Adults through Safe Transitions (BOOST) risk assessment tool was applicable within the UK. METHODS: Patients over 65 readmitted were identified retrospectively via a casenote review. BOOST assessment was applied with 1 point for each risk factor. RESULTS: 324 patients were readmitted (mean age 77 years) with a median of 7 days between discharge and readmission. The median BOOST score was 3 (IQR 2-4) with polypharmacy evident in 88% and prior hospitalisation in 70%. The tool correctly predicted 90% of readmissions using two or more risk factors and 99.1% if one risk factor was included. CONCLUSION: The BOOST assessment tool appears appropriate in predicting readmissions however further analysis is required to determine its precision.

Computing Speed-of-Sound From Ultrasound: User-Agnostic Recovery and a New Benchmark.

OBJECTIVE: Medical ultrasound is one of the most accessible imaging modalities, but is a challenging modality for quantitative parameters comparison across vendors and sonographers. B-Mode imaging, with limited exceptions, provides a map of tissue boundaries; crucially, it does not provide diagnostically relevant physical quantities of the interior of organ domains.This can be remedied: the raw ultrasound signal carries significantly more information than is present in the B-Mode image. Specifically, the ability to recover speed-of-sound and attenuation maps from the raw ultrasound signal transforms the modality into a tissue-property modality. Deep learning was shown to be a viable tool for recovering speed-of-sound maps. A major hold-back towards deployment is the domain transfer problem, i.e., generalizing from simulations to real data. This is due in part to dependence on the (hard-to-calibrate) system response. METHODS: We explore a remedy to the problem of operator-dependent effects on the system response by introducing a novel approach utilizing the phase information of the IQ demodulated signal. RESULTS: We show that the IQ-phase information effectively decouples the operator-dependent system response from the data, significantly improving the stability of speed-of-sound recovery. We also introduce an improvement to the network topology providing faster and improved results to the state-of-the-art. We present the first publicly available benchmark for this problem: a simulated dataset for raw ultrasound plane wave processing. CONCLUSION: The consideration of the phase of the IQ-signals presents a promising appeal to traversing the transfer learning problem, advancing the goal of real-time speed-of-sound imaging.

Principal Uncertainty Quantification With Spatial Correlation for Image Restoration Problems.

Uncertainty quantification for inverse problems in imaging has drawn much attention lately. Existing approaches towards this task define uncertainty regions based on probable values per pixel, while ignoring spatial correlations within the image, resulting in an exaggerated volume of uncertainty. In this paper, we propose PUQ (Principal Uncertainty Quantification) - a novel definition and corresponding analysis of uncertainty regions that takes into account spatial relationships within the image, thus providing reduced volume regions. Using recent advancements in generative models, we derive uncertainty intervals around principal components of the empirical posterior distribution, forming an ambiguity region that guarantees the inclusion of true unseen values with a user-defined confidence probability. To improve computational efficiency and interpretability, we also guarantee the recovery of true unseen values using only a few principal directions, resulting in more informative uncertainty regions. Our approach is verified through experiments on image colorization, super-resolution, and inpainting; its effectiveness is shown through comparison to baseline methods, demonstrating significantly tighter uncertainty regions.

Photon-counting CT urogram: optimal acquisition potential (kV) determination for virtual noncontrast creation.

PURPOSE: To quantitatively and qualitatively compare the degree of iodine removal in the collecting system from PCCT urographic phase-derived virtual noncontrast (VNC) images obtained at 140 kV versus 120 kV. METHODS: A retrospective PACS search identified adult patients (>18 years) who underwent a PCCT urogram for hematuria from 4/2022 to 4/2023 with available urographic phase-derived VNC images in PACS. Tube voltage (120 kV, 140 kV), body mass index, CTDIvol, dose length product (DLP), and size-specific dose estimate (SSDE) were recorded. Hounsfield Unit (HU) in both renal pelvises and the urinary bladder on urographic-derived VNC were recorded. Three radiologists qualitatively assessed the degree of iodine removal (renal pelvis, urinary bladder) and diagnostic confidence for urinary stone detection. Continuous variables were compared for 140 kV versus 120 kV with the Wilcoxon rank sum test. A p < .05 indicated statistical significance. RESULTS: 63 patients (34 male; median (Q1, Q3) age: 30 (26, 34) years; 140 kV/120 kV: 30 patients/33 patients) were included. BMI, CTDIvol, DLP, and SSDE were not different for 140 kV and 120 kV (all p > .05). Median (Q1, Q3) collecting system HU (renal pelvis and bladder) was 0.9 (- 3.6, 4.4) HU at 140 kV and 10.5 (3.6, 26.7) HU at 120 kV (p = .04). Diagnostic confidence for urinary calculi was 4.6 [1.1] at 140 kV and 4.1 [1.4] at 120 kV (p = .005). Diagnostic confidence was 5/5 (all readers) in 82.2% (74/90) at 140 kV and 59.6% (59/99) at 120 kV (p < .001). CONCLUSION: PCCT urographic phase-derived VNC images obtained at 140 kV had better collecting system iodine removal than 120 kV with similar patient radiation exposure. With excellent PCCT urographic phase iodine removal at 140 kV, consideration can be made to utilize a single-phase CT urogram in young patients.

Considerations in the Management of Functional Neurological Disorders in Patients with Hearing Loss.

CASE: Adam is a 14-year-old adolescent boy with hearing loss who presented to the pediatric neurology clinic accompanied by his father for evaluation of new-onset left hand tremor for a duration of 1 month. An American Sign Language interpreter was present and used throughout the visit.Adam has bilateral sensorineural hearing loss related to premature birth at 28 weeks' gestation. He uses sign language and attends a school for the hearing impaired. He has been diagnosed with attention-deficit/hyperactivity disorder (ADHD) and a nonspecific learning disorder. His ADHD symptoms are well controlled with a stimulant medication. He is independent in activities of daily living, and there is no concern for intellectual disability. His father is concerned that Adam may have anxiety, but this has not been evaluated.After careful history, it is found that the tremor was first noted the day after burglars broke into his home and stole precious belongings. Current stressors include difficulties with schoolwork and a strained relationship with an extended family member. There is no family history of tremor.The tremor was intermittent initially, with episodes lasting around 30 minutes. Over time, the tremor became more persistent. Adam is left-handed, and the tremor is now interfering with handwriting, eating, and other fine motor skills. The tremor worsens when Adam is tired or stressed and improves with relaxation. No tremor has been noted in other body parts. Adam denies any other neurological symptoms, including headache, vision changes, or gait abnormalities.On examination, Adam seemed anxious but showed no significant distress and had normal vital signs. His general examination was unremarkable. His neurological examination showed intact cranial nerves, apart from the hearing impairment. He had normal muscle tone, intact strength and coordination, and a normal casual gait. Rhythmic shaking of the left upper extremity was present with action, while maintaining posture. Using specific examination techniques, the examiner was able to alter the rhythm of the tremor, and the tremor was noted to subside when the patient was engaged with the examiner. These findings in addition to signs of suggestibility and variable frequency/direction were consistent with a functional etiology.The diagnosis of a functional neurological disorder manifesting in the form of functional tremor was discussed with the patient and his father with assistance from the sign language interpreter. Counseling regarding management consisting mainly of cognitive behavioral therapy and evaluation of possible coexisting conditions, such as anxiety, was discussed.What factors would you consider in diagnosis and management of functional neurological disorder in a hearing-impaired child/adolescent?

Brief Report: Psychogenic Nonepileptic Events in Pediatric Patients with Autism or Intellectual Disability.

This is a retrospective case series of pediatric patients referred to the psychogenic nonepileptic events clinic (PNEE) who had comorbid diagnoses of autism spectrum disorder (ASD) or intellectual disability (ID). We describe 15 patients, nine with ASD and six with ID who had a telephone visit follow-up at 12 months. There were higher rates of male gender (40%) and comorbid epilepsy (53%) compared to the larger PNEE cohort. Eleven patients were available for follow-up and ten patients had improvement in events or were event-free. We report that patients with ASD or ID can develop PNEE and experience improvement from events.

Functional Neurological Disorders in Children - A Historical Perspective.

The modern diagnosis of functional neurological disorders (FND) describes a diverse set of neurological symptoms without identifiable pathology. The history of FND and its prescientific predecessor, hysteria, is complex and deeply rooted in sexism. Key moments in this history have contributed to the advancement of understanding in functional disorders that neurologists should be aware of. Although pediatric FND has a much less extensive historical literature, there are many parallels between it and the initial interest in hysteria by Charcot and other early modern neurologists. This was followed by waning neurological attention as psychiatrists promoted conversion disorder in the early 1900s. Towards the end of the 20th century, neurologists have taken a renewed, collaborative role with psychiatry and other disciplines to diagnose and study FND.

Novel device for nonsurgical correction of rigid forefoot adduction in children.

Forefoot adduction deformity (FAD) (commonly called metatarsus adductus) is reported as the most common congenital foot deformity in newborns. Early diagnosis and treatment are important in rigid cases, as better outcomes have been reported if treatment was initiated before 9 months of age. While casting and splinting is the current standard of care for nonsurgical management of rigid FAD (RFAD), several orthoses have demonstrated equal benefit. The Universal Neonatal Foot Orthotic (UNFO) brace is below ankle orthosis that provides continuous pressure, thereby correcting the deformity without casting. To the best of our knowledge, UNFO is the first brace that operates below the ankle. The aim of this study was to compare the effectiveness of UNFO shoe to standard serial casting in the treatment of RFAD in infants. Between the years 2012 and 2019 we treated 147 feet (94 patients): 52 using the UNFO shoes and 95 by standard casting and splinting protocol. The treatment groups were compared based on treatment duration, complications, and recurrence of deformity. Mean full-time treatment duration was significantly shorter in the UNFO group, while no significant difference in the total duration of treatment was observed. Similar complication and recurrence rates were demonstrated. In conclusion, treatment with UNFO is equally effective to serial casting. The use of UNFO increases convenience and diminishes social burden, thus providing a distinct advantage over other treatment modalities.

Musculoskeletal and Head Injuries in a Canadian Summer Camp: A Three-Year Surveillance Study.

INTRODUCTION: Residential summer camps are popular among Canadian families. Campers are exposed to new, unfamiliar environments and engage in activities that may pose increased risk of injury. This study identifies the occurrence and management of musculoskeletal and head injuries at a Canadian residential summer camp. METHODS: This study was a three-year prospective observational cohort study, at a six-week Canadian residential summer camp. There were 1,388 residents, consisting of 51,546 camp days (CD). Injury data were collected by residential summer camp staff and confirmed by onsite medical professionals prior to being recorded in a secure database. Injuries were included if it was a musculoskeletal or head injury that occurred while engaged in a camp activity on or offsite, that necessitated medical attention, and that required removal or restriction from their normal camp routine for a minimum of 4 hours. RESULTS: There were 154 injuries, resulting in an incidence of 2.99 injuries per 1000 CD. Injuries were reported during scheduled activities (1.46/1000 CD) and free time (1.20/1000 CD). Sports was the most common activity during which injury occurred in all age groups (1.07/1000 CD), where males were injured twice as often as females. 65% of injuries occurred while under staff supervision. The lower extremity was the most affected body part (1.59/1000 CD). Sprains and strains accounted for 1.69 injuries/1000 CD. 83% of injuries were classified as significant and 89% of injuries were treated on-site. Over-the-counter analgesics were provided in 62% of senior camper injuries and 46% of junior camper and staff injuries. CONCLUSION: Most injuries in the residential camp setting are mild. Ensuring appropriate non-pharmacologic measures in addition to adequate analgesia may help shorten return to play.

Creating a New Set of Milestones for the Clinical Neurophysiology Fellowship.

INTRODUCTION: The Accreditation Council for Graduate Medical Education and the American Board of Psychiatry and Neurology first developed milestones for the clinical neurophysiology (CNP) fellowship in 2015. The milestones provide a comprehensive evaluation of the fellow's development based on six domains of competency. Here, we describe the development of a new set of milestones for CNP fellowship with level 1 as the incoming level, level 4 as the goal for graduation, and level 5 as the aspirational level that may not be achieved. METHODS: Committee members were nominated or volunteered to participate in the milestones update. Milestone development began with the creation of a shared mental model of the ideal skills and knowledge a graduating CNP fellow should attain. RESULTS: The CNP committee met virtually 7 times for a total of 14 meeting hours. Nine Patient Care and five Medical Knowledge milestones evolved from the seven Patient Care and six Medical Knowledge milestones that were in the first iteration. The committee incorporated 11 "Harmonized Milestones" into the revision and a supplemental guide was created. CONCLUSIONS: The revised Accreditation Council for Graduate Medical Education milestones for CNP fellowship contain important updates that program directors should review against their curricula to identify any gaps in learning. Program leadership should take note of two new Patient Care milestones for telemedicine and intraoperative monitoring. Clinical neurophysiology fellowships are not designed to provide level 4 competency across all milestones. The revised milestones should be viewed within the context of an individual program's goals.