Effects of a comprehensive brain computed tomography deep learning model on radiologist detection accuracy.

OBJECTIVES: Non-contrast computed tomography of the brain (NCCTB) is commonly used to detect intracranial pathology but is subject to interpretation errors. Machine learning can augment clinical decision-making and improve NCCTB scan interpretation. This retrospective detection accuracy study assessed the performance of radiologists assisted by a deep learning model and compared the standalone performance of the model with that of unassisted radiologists. METHODS: A deep learning model was trained on 212,484 NCCTB scans drawn from a private radiology group in Australia. Scans from inpatient, outpatient, and emergency settings were included. Scan inclusion criteria were age ≥ 18 years and series slice thickness ≤ 1.5 mm. Thirty-two radiologists reviewed 2848 scans with and without the assistance of the deep learning system and rated their confidence in the presence of each finding using a 7-point scale. Differences in AUC and Matthews correlation coefficient (MCC) were calculated using a ground-truth gold standard. RESULTS: The model demonstrated an average area under the receiver operating characteristic curve (AUC) of 0.93 across 144 NCCTB findings and significantly improved radiologist interpretation performance. Assisted and unassisted radiologists demonstrated an average AUC of 0.79 and 0.73 across 22 grouped parent findings and 0.72 and 0.68 across 189 child findings, respectively. When assisted by the model, radiologist AUC was significantly improved for 91 findings (158 findings were non-inferior), and reading time was significantly reduced. CONCLUSIONS: The assistance of a comprehensive deep learning model significantly improved radiologist detection accuracy across a wide range of clinical findings and demonstrated the potential to improve NCCTB interpretation. CLINICAL RELEVANCE STATEMENT: This study evaluated a comprehensive CT brain deep learning model, which performed strongly, improved the performance of radiologists, and reduced interpretation time. The model may reduce errors, improve efficiency, facilitate triage, and better enable the delivery of timely patient care. KEY POINTS: • This study demonstrated that the use of a comprehensive deep learning system assisted radiologists in the detection of a wide range of abnormalities on non-contrast brain computed tomography scans. • The deep learning model demonstrated an average area under the receiver operating characteristic curve of 0.93 across 144 findings and significantly improved radiologist interpretation performance. • The assistance of the comprehensive deep learning model significantly reduced the time required for radiologists to interpret computed tomography scans of the brain.

Predictors of improvement in quality of life at 12-month follow-up in patients undergoing anterior endoscopic skull base surgery.

BACKGROUND: Patients with pituitary lesions experience decrements in quality of life (QoL) and treatment aims to arrest or improve QoL decline. OBJECTIVE: To detect associations with QoL in trans-nasal endoscopic skull base surgery patients and train supervised learning classifiers to predict QoL improvement at 12 months. METHODS: A supervised learning analysis of a prospective multi-institutional dataset (451 patients) was conducted. QoL was measured using the anterior skull base surgery questionnaire (ASBS). Factors associated with QoL at baseline and at 12-month follow-up were identified using multivariate logistic regression. Multiple supervised learning models were trained to predict postoperative QoL improvement with five-fold cross-validation. RESULTS: ASBS at 12-month follow-up was significantly higher (132.19,SD = 24.87) than preoperative ASBS (121.87,SD = 25.72,p<0.05). High preoperative scores were significantly associated with institution, diabetes and lesions at the planum sphenoidale / tuberculum sella site. Patients with diabetes were five times less likely to report high preoperative QoL. Low preoperative QoL was significantly associated with female gender, a vision-related presentation, diabetes, secreting adenoma and the cavernous sinus site. Top quartile change in postoperative QoL at 12-month follow-up was negatively associated with baseline hypercholesterolemia, acromegaly and intraoperative CSF leak. Positive associations were detected for lesions at the sphenoid sinus site and deficient preoperative endocrine function. AdaBoost, logistic regression and neural network classifiers yielded the strongest predictive performance. CONCLUSION: It was possible to predict postoperative positive change in QoL at 12-month follow-up using perioperative data. Further development and implementation of these models may facilitate improvements in informed consent, treatment decision-making and patient QoL.

Natural Language Processing Applications in the Clinical Neurosciences: A Machine Learning Augmented Systematic Review.

Natural language processing (NLP), a domain of artificial intelligence (AI) that models human language, has been used in medicine to automate diagnostics, detect adverse events, support decision making and predict clinical outcomes. However, applications to the clinical neurosciences appear to be limited. NLP has matured with the implementation of deep transformer models (e.g., XLNet, BERT, T5, and RoBERTa) and transfer learning. The objectives of this study were to (1) systematically review NLP applications in the clinical neurosciences, and (2) explore NLP analysis to facilitate literature synthesis, providing clear examples to demonstrate the potential capabilities of these technologies for a clinical audience. Our NLP analysis consisted of keyword identification, text summarization and document classification. A total of 48 articles met inclusion criteria. NLP has been applied in the clinical neurosciences to facilitate literature synthesis, data extraction, patient identification, automated clinical reporting and outcome prediction. The number of publications applying NLP has increased rapidly over the past five years. Document classifiers trained to differentiate included and excluded articles demonstrated moderate performance (XLNet AUC = 0.66, BERT AUC = 0.59, RoBERTa AUC = 0.62). The T5 transformer model generated acceptable abstract summaries. The application of NLP has the potential to enhance research and practice in the clinical neurosciences.

Effect of a comprehensive deep-learning model on the accuracy of chest x-ray interpretation by radiologists: a retrospective, multireader multicase study.

BACKGROUND: Chest x-rays are widely used in clinical practice; however, interpretation can be hindered by human error and a lack of experienced thoracic radiologists. Deep learning has the potential to improve the accuracy of chest x-ray interpretation. We therefore aimed to assess the accuracy of radiologists with and without the assistance of a deep-learning model. METHODS: In this retrospective study, a deep-learning model was trained on 821 681 images (284 649 patients) from five data sets from Australia, Europe, and the USA. 2568 enriched chest x-ray cases from adult patients (≥16 years) who had at least one frontal chest x-ray were included in the test dataset; cases were representative of inpatient, outpatient, and emergency settings. 20 radiologists reviewed cases with and without the assistance of the deep-learning model with a 3-month washout period. We assessed the change in accuracy of chest x-ray interpretation across 127 clinical findings when the deep-learning model was used as a decision support by calculating area under the receiver operating characteristic curve (AUC) for each radiologist with and without the deep-learning model. We also compared AUCs for the model alone with those of unassisted radiologists. If the lower bound of the adjusted 95% CI of the difference in AUC between the model and the unassisted radiologists was more than -0·05, the model was considered to be non-inferior for that finding. If the lower bound exceeded 0, the model was considered to be superior. FINDINGS: Unassisted radiologists had a macroaveraged AUC of 0·713 (95% CI 0·645-0·785) across the 127 clinical findings, compared with 0·808 (0·763-0·839) when assisted by the model. The deep-learning model statistically significantly improved the classification accuracy of radiologists for 102 (80%) of 127 clinical findings, was statistically non-inferior for 19 (15%) findings, and no findings showed a decrease in accuracy when radiologists used the deep-learning model. Unassisted radiologists had a macroaveraged mean AUC of 0·713 (0·645-0·785) across all findings, compared with 0·957 (0·954-0·959) for the model alone. Model classification alone was significantly more accurate than unassisted radiologists for 117 (94%) of 124 clinical findings predicted by the model and was non-inferior to unassisted radiologists for all other clinical findings. INTERPRETATION: This study shows the potential of a comprehensive deep-learning model to improve chest x-ray interpretation across a large breadth of clinical practice. FUNDING: Annalise.ai.

Machine learning applications to neuroimaging for glioma detection and classification: An artificial intelligence augmented systematic review.

Glioma is the most common primary intraparenchymal tumor of the brain and the 5-year survival rate of high-grade glioma is poor. Magnetic resonance imaging (MRI) is essential for detecting, characterizing and monitoring brain tumors but definitive diagnosis still relies on surgical pathology. Machine learning has been applied to the analysis of MRI data in glioma research and has the potential to change clinical practice and improve patient outcomes. This systematic review synthesizes and analyzes the current state of machine learning applications to glioma MRI data and explores the use of machine learning for systematic review automation. Various datapoints were extracted from the 153 studies that met inclusion criteria and analyzed. Natural language processing (NLP) analysis involved keyword extraction, topic modeling and document classification. Machine learning has been applied to tumor grading and diagnosis, tumor segmentation, non-invasive genomic biomarker identification, detection of progression and patient survival prediction. Model performance was generally strong (AUC = 0.87 ± 0.09; sensitivity = 0.87 ± 0.10; specificity = 0.0.86 ± 0.10; precision = 0.88 ± 0.11). Convolutional neural network, support vector machine and random forest algorithms were top performers. Deep learning document classifiers yielded acceptable performance (mean 5-fold cross-validation AUC = 0.71). Machine learning tools and data resources were synthesized and summarized to facilitate future research. Machine learning has been widely applied to the processing of MRI data in glioma research and has demonstrated substantial utility. NLP and transfer learning resources enabled the successful development of a replicable method for automating the systematic review article screening process, which has potential for shortening the time from discovery to clinical application in medicine.

Machine learning applications to clinical decision support in neurosurgery: an artificial intelligence augmented systematic review.

Machine learning (ML) involves algorithms learning patterns in large, complex datasets to predict and classify. Algorithms include neural networks (NN), logistic regression (LR), and support vector machines (SVM). ML may generate substantial improvements in neurosurgery. This systematic review assessed the current state of neurosurgical ML applications and the performance of algorithms applied. Our systematic search strategy yielded 6866 results, 70 of which met inclusion criteria. Performance statistics analyzed included area under the receiver operating characteristics curve (AUC), accuracy, sensitivity, and specificity. Natural language processing (NLP) was used to model topics across the corpus and to identify keywords within surgical subspecialties. ML applications were heterogeneous. The densest cluster of studies focused on preoperative evaluation, planning, and outcome prediction in spine surgery. The main algorithms applied were NN, LR, and SVM. Input and output features varied widely and were listed to facilitate future research. The accuracy (F(2,19) = 6.56, p < 0.01) and specificity (F(2,16) = 5.57, p < 0.01) of NN, LR, and SVM differed significantly. NN algorithms demonstrated significantly higher accuracy than LR. SVM demonstrated significantly higher specificity than LR. We found no significant difference between NN, LR, and SVM AUC and sensitivity. NLP topic modeling reached maximum coherence at seven topics, which were defined by modeling approach, surgery type, and pathology themes. Keywords captured research foci within surgical domains. ML technology accurately predicts outcomes and facilitates clinical decision-making in neurosurgery. NNs frequently outperformed other algorithms on supervised learning tasks. This study identified gaps in the literature and opportunities for future neurosurgical ML research.

Investigating Risk Factors and Predicting Complications in Deep Brain Stimulation Surgery with Machine Learning Algorithms.

BACKGROUND: Deep brain stimulation (DBS) surgery is an option for patients experiencing medically resistant neurologic symptoms. DBS complications are rare; finding significant predictors requires a large number of surgeries. Machine learning algorithms may be used to effectively predict these outcomes. The aims of this study were to 1) investigate preoperative clinical risk factors and 2) build machine learning models to predict adverse outcomes. METHODS: This multicenter registry collected clinical and demographic characteristics of patients undergoing DBS surgery (n = 501) and tabulated occurrence of complications. Logistic regression was used to evaluate risk factors. Supervised learning algorithms were trained and validated on 70% and 30%, respectively, of both oversampled and original registry data. Performance was evaluated using area under the receiver operating characteristics curve (AUC), sensitivity, specificity, and accuracy. RESULTS: Logistic regression showed that the risk of complication was related to the operating institution in which the surgery was performed (odds ratio [OR] = 0.44, confidence interval [CI] = 0.25-0.78), body mass index (OR = 0.94, CI = 0.89-0.99), and diabetes (OR = 2.33, CI = 1.18-4.60). Patients with diabetes were almost 3× more likely to return to the operating room (OR = 2.78, CI = 1.31-5.88). Patients with a history of smoking were 4× more likely to experience postoperative infection (OR = 4.20, CI = 1.21-14.61). Supervised learning algorithms demonstrated high discrimination performance when predicting any complication (AUC = 0.86), a complication within 12 months (AUC = 0.91), return to the operating room (AUC = 0.88), and infection (AUC = 0.97). Age, body mass index, procedure side, gender, and a diagnosis of Parkinson disease were influential features. CONCLUSIONS: Multiple significant complication risk factors were identified, and supervised learning algorithms effectively predicted adverse outcomes in DBS surgery.

Biceps Tenotomy Versus Tenodesis in Active Patients Younger Than 55 Years: Is There a Difference in Strength and Outcomes?

BACKGROUND: Proximal biceps pathology is a significant factor in shoulder pain. Surgical treatment options include biceps tenotomy and subpectoral biceps tenodesis. Tenotomy is a simple procedure, but it may produce visible deformity, subjective cramping, or loss of supination strength. Tenodesis is a comparatively technical procedure involving a longer recovery, but it has been hypothesized to achieve better outcomes in younger active patients (<55 years). HYPOTHESIS: This study investigated the outcomes of younger patients who underwent either a biceps tenotomy or tenodesis as part of treatment for shoulder pain. The hypothesis was that, apart from cosmetic deformity, there will be no difference in outcome between the 2 treatment options. STUDY DESIGN: Cohort study; Level of evidence, 3. METHODS: Isometric strength and endurance testing of operative and nonoperative shoulders for forearm supination (FS) and elbow flexion (EF) were tested utilizing an isometric dynamometer. Objective physical assessment was also performed. Subjective outcomes using the modified American Shoulder and Elbow Surgeons score (ASES); Disability of the Arm, Shoulder, and Hand (DASH); visual analog scale (VAS); and perceived biceps symptoms were collected. RESULTS: A total of 42 patients (22 tenotomy, 20 tenodesis) with an average follow-up of 3.3 years were studied. The average age at follow-up was 49.9 years. Thirty-five percent (7/20) of tenotomy patients exhibited a "Popeye" deformity, compared with 18.2% (4/22) of tenodesis patients. Strength prior to fatiguing exercise was similar between tenodesis and tenotomy for FS (6.9 vs 7.3 lbs; P < .05), EF in neutral (35.4 vs 35.4 lbs), and EF in supination (33.8 vs 34.2 lbs). Strength was not significantly different between groups for isometric strength and endurance measures. Subjective functional outcome measured by the DASH, ASES, and VAS scores were similar between groups. Frequency of complaints of cramping was higher in the tenotomy group (4/20 vs 1/22), and complaints of pain were higher in the tenodesis group (11/22 vs 5/20). CONCLUSION: Despite increased demands and activity placed on biceps function in a younger population, this study showed no differences in functional and subjective outcome measurements. The choice between biceps tenotomy and tenodesis for pathology of the proximal biceps tendon can continue to be based on surgeon and patient preference.

The administration of rescue medication to children with prolonged acute convulsive seizures in a non-hospital setting: an exploratory survey of healthcare professionals' perspectives.

UNLABELLED: Timely administration of rescue medication for prolonged convulsive seizures lasting more than 5 min is necessary to prevent progression to status epilepticus in children. The Practices in Emergency and Rescue Medication for Epilepsy Managed with Community Administered Therapy (PERFECT™) initiative was set up to gain a better understanding of how prolonged convulsive seizures in children are managed when they occur outside of the hospital. We present the findings from an exploratory telephone survey of 128 healthcare professionals (HCPs) (85 pediatric neurologists and neurologists, 28 community pediatricians, and 15 epilepsy nurses) from six EU countries, conducted as part of the PERFECT™ initiative. Among HCPs, there was greater awareness of local protocols and lesser awareness of national or international guidelines. HCPs were not very aware of how prolonged convulsive seizures were managed outside of the hospital and had few professional links to schools or other settings where these seizures were most likely to occur. Approximately one third believed that lack of confidence and fear of liability were barriers to caregivers administering treatment in schools, as was insufficient training of caregivers in the wider community on the management of prolonged convulsive seizures. CONCLUSION: Results of this HCP survey have identified several clear gaps that need to be addressed: clearer guidance that spans all settings of care, greater dissemination of such guidelines across the chain of care, more open communication and better links between HCPs and schools, and systematic training of all relevant caregivers on the appropriate management of prolonged convulsive seizures.

Are we there yet? A journey of health reform in Australia.

• Five years on from the establishment of the National Health and Hospitals Reform Commission, it is timely to review the context for reform and some of the actions taken to date, and to highlight remaining areas of concern and priority. • The Commission's final report was released in July 2009 and presented 123 recommendations organised under four reform themes: Taking responsibility: individual and collective action to build good health and wellbeing - by people, families, communities, health professionals, employers, health funders and governments Connecting care: comprehensive care for people over their lifetime Facing inequities: recognise and tackle the causes and impacts of health inequities Driving quality performance: leadership and systems to achieve best use of people, resources and evolving knowledge. • Overall, the Australian Government's response to the Commission's report has been very positive, but challenges remain in some key areas: Financial sustainability and the vertical fiscal imbalance between the federal and state governments Getting the best value from the health dollar by reducing inefficiency and waste and using value-based purchasing across the public and private health sectors National leadership across the system as a whole Getting the right care in the right place at the right time Health is about more than health care - increasing focus on prevention and recognising and tackling the broader social determinants of health.

Are we failing to provide adequate rescue medication to children at risk of prolonged convulsive seizures in schools?

OBJECTIVE: This paper explores the issues that arise from the discussion of administering rescue medication to children who experience prolonged convulsive seizures in mainstream schools in the UK. SITUATION ANALYSIS: Current guidelines recommend immediate treatment of children with such seizures (defined as seizures lasting more than 5 min) to prevent progression to status epilepticus and neurological morbidity. As children are unconscious during prolonged convulsive seizures, whether or not they receive their treatment in time depends on the presence of a teacher or other member of staff trained and able to administer rescue medication. However, it is thought that the situation varies between schools and depends mainly on the goodwill and resources available locally. RECOMMENDATIONS: A more systematic response is needed to ensure that children receive rescue medication regardless of where their seizure occurs. Possible ways forward include: greater use of training resources for schools available from epilepsy voluntary sector organisations; consistent, practical information to schools; transparent guidance outlining a clear care pathway from the hospital to the school; and implementation and adherence to each child's individual healthcare plan. IMPLICATIONS: Children requiring emergency treatment for prolonged convulsive seizures during school hours test the goals of integrated, person-centred care as well as joined-up working to which the National Health Service (NHS) aspires. As changes to the NHS come into play and local services become reconfigured, every effort should be made to take account of the particular needs of this vulnerable group of children within broader efforts to improve the quality of paediatric epilepsy services overall.

The administration of rescue medication to children with prolonged acute convulsive seizures in the community: what happens in practice?

This paper presents the findings of a review of existing clinical and non-clinical guidance on the management of children with prolonged acute convulsive seizures (PCS) and the administration of rescue medication in community settings. Findings are based on desk- and web-based research in 6 countries. Published clinical guidelines are mostly limited to the hospital setting and offer few explicit recommendations for community settings. Non-clinical guidance on the management of medicines at school exists at the national or regional level in all 6 countries, however rescue epilepsy medication is often not mentioned specifically. Existing legal frameworks are vague and open to interpretation. As a result, whether a child receives rescue medication at school depends primarily on the availability of a willing teacher who accepts responsibility for administering it to that child during school hours. Comprehensive guidelines are clearly needed that provide practical guidance to ensure that children with PCS are treated as quickly as possible in all community settings. Recommendations for future action include: providing clearer information on PCS and rescue medication to parents and schools; putting in place an individual healthcare plan for every child with a history of PCS at his or her school; collecting more empirical data to gain a better understanding of the experience of children with PCS at school, their parents and teachers; and finally, implementing systematic training for all carers of children with PCS. The epilepsy specialist may play an important role in ensuring that these recommendations are put into place for their patients.

Metabolism and performance during extended high-intensity intermittent exercise after consumption of low- and high-glycaemic index pre-exercise meals.

The metabolic and performance benefits of prior consumption of low-glycaemic index (GI) meals v. high-GI meals were determined in extended high-intensity intermittent exercise. Participants (ten males and four females, aged 25·8 (sd 7·3) years) completed two testing days (each consisting of back-to-back 90-min intermittent high-intensity treadmill running protocols separated by 3 h) spaced by at least 7 d. Using a randomised counterbalanced cross-over design, low-GI, lentil-based meals (GI about 42) or high-GI, potato-based meals (GI about 78) matched for energy value were consumed 2 h before, and within 1 h after, the first exercise session. Performance was measured by the distance covered during five 1-min sprints (separated by 2·5 min walking) at the end of each exercise session. Peak postprandial blood glucose was higher by 30·8 % in the high-GI trial compared with the low-GI trial, as was insulin (P = 0·039 and P = 0·003, respectively). Carbohydrate oxidation was lower by 5·5 % during the low-GI trials compared with the high-GI trials at the start of the first exercise session (P < 0·05). Blood lactate was significantly higher (6·1 v. 2·6 mmol/l; P = 0·019) and blood glucose significantly lower (4·8 v. 5·4 mmol/l; P = 0·039) at the end of the second exercise session during the high-GI trial compared with the low-GI trial. Sprint distance was not significantly different between conditions. A low-GI meal improved the metabolic profile before and during extended high-intensity intermittent exercise, but did not affect performance. Improvements in metabolic responses when consuming low-GI meals before exercise may be beneficial to the long-term health of athletes.

The impact of population-based disease management services for selected chronic conditions: the Costs to Australian Private Insurance--Coaching Health (CAPICHe) study protocol.

BACKGROUND: Recent evidence from a large scale trial conducted in the United States indicates that enhancing shared decision-making and improving knowledge, self-management, and provider communication skills to at-risk patients can reduce health costs and utilisation of healthcare resources. Although this trial has provided a significant advancement in the evidence base for disease management programs it is still left for such results to be replicated and/or generalised for populations in other countries and other healthcare environments. This trial responds to the limited analyses on the effectiveness of providing chronic disease management services through telephone health coaching in Australia. The size of this trial and it's assessment of cost utility with respect to potentially preventable hospitalisations adds significantly to the body of knowledge to support policy and investment decisions in Australia as well as to the international debate regarding the effect of disease management programs on financial outcomes. METHODS: Intention to treat study applying a prospective randomised design comparing usual care with extensive outreach to encourage use of telephone health coaching for those people identified from a risk scoring algorithm as having a higher likelihood of future health costs. The trial population has been limited to people with one or more of the following selected chronic conditions: namely, low back pain, diabetes, coronary artery disease, heart failure, and chronic obstructive pulmonary disease. This trial will enrol at least 64,835 sourced from the approximately 3 million Bupa Australia private health insured members located across Australia. The primary outcome will be the total (non-maternity) cost per member as reported to the private health insurer (i.e. charged to the insurer) 12 months following entry into the trial for each person. Study recruitment will be completed in early 2012 and the results will be available in late 2013. DISCUSSION: If positive, CAPICHe will represent a potentially cost-effective strategy to improve health outcomes in higher risk individuals with a chronic condition, in a private health insurance setting. TRIAL REGISTRATION: Australian New Zealand Clinical Trials Registry reference: ACTRN12611000580976.