Correlating natural language processing and automated speech analysis with clinician assessment to quantify speech-language changes in mild cognitive impairment and Alzheimer's dementia.

BACKGROUND: Language impairment is an important marker of neurodegenerative disorders. Despite this, there is no universal system of terminology used to describe these impairments and large inter-rater variability can exist between clinicians assessing language. The use of natural language processing (NLP) and automated speech analysis (ASA) is emerging as a novel and potentially more objective method to assess language in individuals with mild cognitive impairment (MCI) and Alzheimer's dementia (AD). No studies have analyzed how variables extracted through NLP and ASA might also be correlated to language impairments identified by a clinician. METHODS: Audio recordings (n=30) from participants with AD, MCI, and controls were rated by clinicians for word-finding difficulty, incoherence, perseveration, and errors in speech. Speech recordings were also transcribed, and linguistic and acoustic variables were extracted through NLP and ASA. Correlations between clinician-rated speech characteristics and the variables were compared using Spearman's correlation. Exploratory factor analysis was applied to find common factors between variables for each speech characteristic. RESULTS: Clinician agreement was high in three of the four speech characteristics: word-finding difficulty (ICC = 0.92, p<0.001), incoherence (ICC = 0.91, p<0.001), and perseveration (ICC = 0.88, p<0.001). Word-finding difficulty and incoherence were useful constructs at distinguishing MCI and AD from controls, while perseveration and speech errors were less relevant. Word-finding difficulty as a construct was explained by three factors, including number and duration of pauses, word duration, and syntactic complexity. Incoherence was explained by two factors, including increased average word duration, use of past tense, and changes in age of acquisition, and more negative valence. CONCLUSIONS: Variables extracted through automated acoustic and linguistic analysis of MCI and AD speech were significantly correlated with clinician ratings of speech and language characteristics. Our results suggest that correlating NLP and ASA with clinician observations is an objective and novel approach to measuring speech and language changes in neurodegenerative disorders.

Dural Tears During Lumbar Spinal Endoscopy: Surgeon Skill, Training, Incidence, Risk Factors, and Management.

BACKGROUND: Incidental dural tears during lumbar endoscopy can be challenging to manage. There is limited literature on their appropriate management, risk factors, and the clinical consequences of this typically uncommon complication. MATERIALS AND METHODS: To improve the statistical power of studying durotomy with lumbar endoscopy, we performed a retrospective survey study among endoscopic spine surgeons by email and chat groups on social media networks, including WhatsApp and WeChat. Descriptive and correlative statistics were done on the surgeons' recorded responses to multiple-choice questions. Surgeons were asked about their clinical experience with spinal endoscopy, training background, the types of lumbar endoscopic decompression they perform by approach, the decompression instruments they use, and incidental durotomy incidence with routine lumbar endoscopy. RESULTS: There were 689 dural tears in 64 470 lumbar endoscopies, resulting in an incidental durotomy incidence of 1.07%. Seventy percent of the durotomies were reported by 20.4% of the surgeons. Eliminating these 19 outlier surgeons yielded an adjusted durotomy rate of 0.32. Endoscopic stenosis decompression (54.8%; P < .0001), rather than endoscopic discectomy (44.1%; 41/93), was significantly more associated with durotomy. Medium-sized dural tears (1-10 mm) were the most common (52.2%; 48/93). Small pinhole durotomies (less than 1 mm) were the second most common type (46.7%; 43/93). Rootlet herniations were seen by 46.2% (43/93) of responding surgeons. The posterior dural sac injury during the interlaminar approach (57%; 53/93) occurred more frequently than traversing nerve-root injuries (31.2%) or anterior dural sac (23.7%; 22/93). Exiting nerve-root injuries (10.8%;10/93) were less common. Over half of surgeons did not attempt any repair or closure (52.2%; 47/90). Forty percent (36/90) used sealants. Only 7.8% (7/90) of surgeons attempted an endoscopic repair or sutures (11.1%; 10/90). DuralSeal was the most commonly used brand of commercially available sealant used (42.7%; 35/82). However, other sealants such as Tisseal (15.9%; 13/82), Evicel (2.4%2/82), and additional no-brand sealants (38; 32/82) were also used. Nearly half of the patients (48.3%; 43/89) were treated with 24-48 hours of bed rest. The majority of participating surgeons (64%; 57/89) reported that the long-term outcome was unaffected. Only 18% of surgeons reported having seen the development of a postoperative cerebrospinal fluid (CSF)-fistula (18%;16/89). However, the absolute incidence of CSF fistula was only 0.025% (16/64 470). Severe radiculopathy with dysesthesia; sensory loss; and motor weakness in association with an incidental durotomy were reported by 12.4% (11/89), 3.4% (3/89), and 2.2% (2/89) of surgeons, respectively. CONCLUSIONS: The incidence of dural tears with lumbar endoscopy is about 1%. The incidence of durotomy is higher with the use of power drills and the interlaminar approach. Stenosis decompression that typically requires the more aggressive use of these power instruments has a slightly higher incidence of dural tears than does endoscopic decompression for a herniated disc. Most dural tears are small and can be successfully managed with mechanical compression with Gelfoam and sealants. Two-thirds of patients with incidental dural tears had an entirely uneventful postoperative course. The remaining one-third of patients may develop a persistent CSF leak, radiculopathy with dysesthesia, sensory loss, or motor function loss. Patients should be educated preoperatively and reassured. LEVEL OF EVIDENCE: 3.

A Primer on Usability Assessment Approaches for Health-Related Applications of Virtual Reality.

Health-related virtual reality (VR) applications for patient treatment, rehabilitation, and medical professional training are on the rise. However, there is little guidance on how to select and perform usability evaluations for VR health interventions compared to the supports that exist for other digital health technologies. The purpose of this viewpoint paper is to present an introductory summary of various usability testing approaches or methods that can be used for VR applications. Along with an overview of each, a list of resources is provided for readers to obtain additionally relevant information. Six categories of VR usability evaluations are described using a previously developed classification taxonomy specific to VR environments: (1) cognitive or task walkthrough, (2) graphical evaluation, (3) post hoc questionnaires or interviews, (4) physical performance evaluation, (5) user interface evaluation, and (6) heuristic evaluation. Given the growth of VR in health care, rigorous evaluation and usability testing is crucial in the development and implementation of novel VR interventions. The approaches outlined in this paper provide a starting point for conducting usability assessments for health-related VR applications; however, there is a need to also move beyond these to adopt those from the gaming industry, where assessments for both usability and user experience are routinely conducted.

Assessment of two immunodepletion methods: off-target effects and variations in immunodepletion efficiency may confound plasma proteomics.

Immunodepletion of abundant plasma proteins increases the depth of proteome penetration by mass spectrometry. However, the nature and extent of immunodepletion and the effect of off-target depletion on the quantitative comparison of the residual proteins have not been critically addressed. We performed mass spectrometry label-free quantitation to determine which proteins were immunodepleted and by how much. Two immunodepletion resins were compared: Qproteome (Qiagen) which removes albumin+immunoglobulins and Seppro IgY14+SuperMix (Sigma-Aldrich) which removes 14 target proteins plus a number of unidentified proteins. Plasma collected by P100 proteomic plasma collection tubes (BD) from 20 human subjects was individually immunodepleted to minimize potential variability, prior to pooling. The abundant proteins were quantified better when using only albumin+immunoglobulins removal (Qproteome), while lower abundance proteins were evaluated better using exhaustive immunodepletion (Seppro IgY14+SuperMix). The latter resin removed at least 155 proteins, 38% of the plasma proteome in protein number and 94% of plasma protein in mass. The depth of immunodepletion likely accounts for the effectiveness of this resin in revealing low abundance proteins. However, the more profound immunodepletion achieved with the IgY14+SuperMix may lead to false-positive fold-changes between comparison groups if the reproducibility and efficiency of the depletion of a given protein are not considered.

Enzymatic mutation detection technologies.

Mutation is as necessary for life as fidelity is in DNA replication. The study of mutations reveals the normal functions of genes, messages, proteins, the causes of many diseases, and the variability of responses among individuals. Indeed, recent mutations that have not yet become polymorphisms are often deleterious and pertinent to the disease history of afflicted individuals. This review discusses the principles behind a variety of methods for the detection of mutations and factors that should be considered in future methods design. One enzymatic approach in particular using orthologs of the CEL I nuclease that show high specificity for all mismatches, appears to be easy and robust. Further developments of this and other methods will allow mutation detection to become an integral component of individualized medicine.