Evaluating the impact of artificial intelligence-assisted image analysis on the diagnostic accuracy of front-line clinicians in detecting fractures on plain X-rays (FRACT-AI): protocol for a prospective observational study.

INTRODUCTION: Missed fractures are the most frequent diagnostic error attributed to clinicians in UK emergency departments and a significant cause of patient morbidity. Recently, advances in computer vision have led to artificial intelligence (AI)-enhanced model developments, which can support clinicians in the detection of fractures. Previous research has shown these models to have promising effects on diagnostic performance, but their impact on the diagnostic accuracy of clinicians in the National Health Service (NHS) setting has not yet been fully evaluated. METHODS AND ANALYSIS: A dataset of 500 plain radiographs derived from Oxford University Hospitals (OUH) NHS Foundation Trust will be collated to include all bones except the skull, facial bones and cervical spine. The dataset will be split evenly between radiographs showing one or more fractures and those without. The reference ground truth for each image will be established through independent review by two senior musculoskeletal radiologists. A third senior radiologist will resolve disagreements between two primary radiologists. The dataset will be analysed by a commercially available AI tool, BoneView (Gleamer, Paris, France), and its accuracy for detecting fractures will be determined with reference to the ground truth diagnosis. We will undertake a multiple case multiple reader study in which clinicians interpret all images without AI support, then repeat the process with access to AI algorithm output following a 4-week washout. 18 clinicians will be recruited as readers from four hospitals in England, from six distinct clinical groups, each with three levels of seniority (early-stage, mid-stage and later-stage career). Changes in the accuracy, confidence and speed of reporting will be compared with and without AI support. Readers will use a secure web-based DICOM (Digital Imaging and Communications in Medicine) viewer (www.raiqc.com), allowing radiograph viewing and abnormality identification. Pooled analyses will be reported for overall reader performance as well as for subgroups including clinical role, level of seniority, pathological finding and difficulty of image. ETHICS AND DISSEMINATION: The study has been approved by the UK Healthcare Research Authority (IRAS 310995, approved on 13 December 2022). The use of anonymised retrospective radiographs has been authorised by OUH NHS Foundation Trust. The results will be presented at relevant conferences and published in a peer-reviewed journal. TRIAL REGISTRATION NUMBERS: This study is registered with ISRCTN (ISRCTN19562541) and ClinicalTrials.gov (NCT06130397). The paper reports the results of a substudy of STEDI2 (Simulation Training for Emergency Department Imaging Phase 2).

Evaluation of the impact of artificial intelligence-assisted image interpretation on the diagnostic performance of clinicians in identifying pneumothoraces on plain chest X-ray: a multi-case multi-reader study.

BACKGROUND: Artificial intelligence (AI)-assisted image interpretation is a fast-developing area of clinical innovation. Most research to date has focused on the performance of AI-assisted algorithms in comparison with that of radiologists rather than evaluating the algorithms' impact on the clinicians who often undertake initial image interpretation in routine clinical practice. This study assessed the impact of AI-assisted image interpretation on the diagnostic performance of frontline acute care clinicians for the detection of pneumothoraces (PTX). METHODS: A multicentre blinded multi-case multi-reader study was conducted between October 2021 and January 2022. The online study recruited 18 clinician readers from six different clinical specialties, with differing levels of seniority, across four English hospitals. The study included 395 plain CXR images, 189 positive for PTX and 206 negative. The reference standard was the consensus opinion of two thoracic radiologists with a third acting as arbitrator. General Electric Healthcare Critical Care Suite (GEHC CCS) PTX algorithm was applied to the final dataset. Readers individually interpreted the dataset without AI assistance, recording the presence or absence of a PTX and a confidence rating. Following a 'washout' period, this process was repeated including the AI output. RESULTS: Analysis of the performance of the algorithm for detecting or ruling out a PTX revealed an overall AUROC of 0.939. Overall reader sensitivity increased by 11.4% (95% CI 4.8, 18.0, p=0.002) from 66.8% (95% CI 57.3, 76.2) unaided to 78.1% aided (95% CI 72.2, 84.0, p=0.002), specificity 93.9% (95% CI 90.9, 97.0) without AI to 95.8% (95% CI 93.7, 97.9, p=0.247). The junior reader subgroup showed the largest improvement at 21.7% (95% CI 10.9, 32.6), increasing from 56.0% (95% CI 37.7, 74.3) to 77.7% (95% CI 65.8, 89.7, p<0.01). CONCLUSION: The study indicates that AI-assisted image interpretation significantly enhances the diagnostic accuracy of clinicians in detecting PTX, particularly benefiting less experienced practitioners. While overall interpretation time remained unchanged, the use of AI improved diagnostic confidence and sensitivity, especially among junior clinicians. These findings underscore the potential of AI to support less skilled clinicians in acute care settings.

Imaging Response to Treatment in Rheumatology.

This article highlights the crucial role of various imaging techniques in the diagnosis and monitoring of rheumatologic diseases. It provides an overview of the different modalities available for imaging rheumatic diseases, the disease processes they are able to demonstrate, and their utility in the monitoring response to therapy. It emphasizes the need for a multifaceted approach that combines radiography, ultrasound, MR imaging, and PET imaging to gain a comprehensive understanding of disease progression and treatment response. Standardized grading systems along with quantitative imaging techniques are playing an increasing role in monitoring disease activity and assessing response to therapy.

Assessing the effectiveness of artificial intelligence (AI) in prioritising CT head interpretation: study protocol for a stepped-wedge cluster randomised trial (ACCEPT-AI).

INTRODUCTION: Diagnostic imaging is vital in emergency departments (EDs). Accessibility and reporting impacts ED workflow and patient care. With radiology workforce shortages, reporting capacity is limited, leading to image interpretation delays. Turnaround times for image reporting are an ED bottleneck. Artificial intelligence (AI) algorithms can improve productivity, efficiency and accuracy in diagnostic radiology, contingent on their clinical efficacy. This includes positively impacting patient care and improving clinical workflow. The ACCEPT-AI study will evaluate Qure.ai's qER software in identifying and prioritising patients with critical findings from AI analysis of non-contrast head CT (NCCT) scans. METHODS AND ANALYSIS: This is a multicentre trial, spanning four diverse sites, over 13 months. It will include all individuals above the age of 18 years who present to the ED, referred for an NCCT. The project will be divided into three consecutive phases (pre-implementation, implementation and post-implementation of the qER solution) in a stepped-wedge design to control for adoption bias and adjust for time-based changes in the background patient characteristics. Pre-implementation involves baseline data for standard care to support the primary and secondary outcomes. The implementation phase includes staff training and qER solution threshold adjustments in detecting target abnormalities adjusted, if necessary. The post-implementation phase will introduce a notification (prioritised flag) in the radiology information system. The radiologist can choose to agree with the qER findings or ignore it according to their clinical judgement before writing and signing off the report. Non-qER processed scans will be handled as per standard care. ETHICS AND DISSEMINATION: The study will be conducted in accordance with the principles of Good Clinical Practice. The protocol was approved by the Research Ethics Committee of East Midlands (Leicester Central), in May 2023 (REC (Research Ethics Committee) 23/EM/0108). Results will be published in peer-reviewed journals and disseminated in scientific findings (ClinicalTrials.gov: NCT06027411) TRIAL REGISTRATION NUMBER: NCT06027411.

Visual Field Deficits in Albinism in Comparison to Idiopathic Infantile Nystagmus.

Purpose: This is the first systematic comparison of visual field (VF) deficits in people with albinism (PwA) and idiopathic infantile nystagmus (PwIIN) using static perimetry. We also compare best-corrected visual acuity (BCVA) and optical coherence tomography measures of the fovea, parafovea, and circumpapillary retinal nerve fiber layer in PwA. Methods: VF testing was performed on 62 PwA and 36 PwIIN using a Humphrey Field Analyzer (SITA FAST 24-2). Mean detection thresholds for each eye were calculated, along with quadrants and central measures. Retinal layers were manually segmented in the macular region. Results: Mean detection thresholds were significantly lower than normative values for PwA (-3.10 ± 1.67 dB, P << 0.0001) and PwIIN (-1.70 ± 1.54 dB, P < 0.0001). Mean detection thresholds were significantly lower in PwA compared to PwIIN (P < 0.0001) and significantly worse for left compared to right eyes in PwA (P = 0.0002) but not in PwIIN (P = 0.37). In PwA, the superior nasal VF was significantly worse than other quadrants (P < 0.05). PwIIN appeared to show a mild relative arcuate scotoma. In PwA, central detection thresholds were correlated with foveal changes in the inner and outer retina. VF was strongly correlated to BCVA in both groups. Conclusions: Clear peripheral and central VF deficits exist in PwA and PwIIN, and static VF results need to be interpreted with caution clinically. Since PwA exhibit considerably lower detection thresholds compared to PwIIN, VF defects are unlikely to be due to nystagmus in PwA. In addition to horizontal VF asymmetry, PwA exhibit both vertical and interocular asymmetries, which needs further exploration.

Phenotypic Features Determining Visual Acuity in Albinism and the Role of Amblyogenic Factors.

Albinism is a spectrum disorder causing foveal hypoplasia, nystagmus, and hypopigmentation of the iris and fundus along with other visual deficits, which can all impact vision. Albinism is also associated with amblyogenic factors which could affect monocular visual acuity. The foveal appearance in albinism can range from mild foveal hypoplasia to that which is indistinguishable from the peripheral retina. The appearance can be quickly and easily graded using the Leicester Grading System in the clinic. However, interquartile ranges of 0.3 logMAR for the grades associated with albinism limit the accuracy of the grading system in predicting vision. Here, we discuss the potential role of nystagmus presenting evidence that it may not be a major source of variability in the prediction of visual acuity. We also show that interocular differences in visual acuity are low in albinism despite high levels of amblyogenic factors indicating that active suppression of vision in one eye in albinism is uncommon.

AI assisted reader evaluation in acute CT head interpretation (AI-REACT): protocol for a multireader multicase study.

INTRODUCTION: A non-contrast CT head scan (NCCTH) is the most common cross-sectional imaging investigation requested in the emergency department. Advances in computer vision have led to development of several artificial intelligence (AI) tools to detect abnormalities on NCCTH. These tools are intended to provide clinical decision support for clinicians, rather than stand-alone diagnostic devices. However, validation studies mostly compare AI performance against radiologists, and there is relative paucity of evidence on the impact of AI assistance on other healthcare staff who review NCCTH in their daily clinical practice. METHODS AND ANALYSIS: A retrospective data set of 150 NCCTH will be compiled, to include 60 control cases and 90 cases with intracranial haemorrhage, hypodensities suggestive of infarct, midline shift, mass effect or skull fracture. The intracranial haemorrhage cases will be subclassified into extradural, subdural, subarachnoid, intraparenchymal and intraventricular. 30 readers will be recruited across four National Health Service (NHS) trusts including 10 general radiologists, 15 emergency medicine clinicians and 5 CT radiographers of varying experience. Readers will interpret each scan first without, then with, the assistance of the qER EU 2.0 AI tool, with an intervening 2-week washout period. Using a panel of neuroradiologists as ground truth, the stand-alone performance of qER will be assessed, and its impact on the readers' performance will be analysed as change in accuracy (area under the curve), median review time per scan and self-reported diagnostic confidence. Subgroup analyses will be performed by reader professional group, reader seniority, pathological finding, and neuroradiologist-rated difficulty. ETHICS AND DISSEMINATION: The study has been approved by the UK Healthcare Research Authority (IRAS 310995, approved 13 December 2022). The use of anonymised retrospective NCCTH has been authorised by Oxford University Hospitals. The results will be presented at relevant conferences and published in a peer-reviewed journal. TRIAL REGISTRATION NUMBER: NCT06018545.

CHIASM-Net: Artificial Intelligence-Based Direct Identification of Chiasmal Abnormalities in Albinism.

Purpose: Albinism is a congenital disorder affecting pigmentation levels, structure, and function of the visual system. The identification of anatomical changes typical for people with albinism (PWA), such as optic chiasm malformations, could become an important component of diagnostics. Here, we tested an application of convolutional neural networks (CNNs) for this purpose. Methods: We established and evaluated a CNN, referred to as CHIASM-Net, for the detection of chiasmal malformations from anatomic magnetic resonance (MR) images of the brain. CHIASM-Net, composed of encoding and classification modules, was developed using MR images of controls (n = 1708) and PWA (n = 32). Evaluation involved 8-fold cross validation involving accuracy, precision, recall, and F1-score metrics and was performed on a subset of controls and PWA samples excluded from the training. In addition to quantitative metrics, we used Explainable AI (XAI) methods that granted insights into factors driving the predictions of CHIASM-Net. Results: The results for the scenario indicated an accuracy of 85 ± 14%, precision of 90 ± 14% and recall of 81 ± 18%. XAI methods revealed that the predictions of CHIASM-Net are driven by optic-chiasm white matter and by the optic tracts. Conclusions: CHIASM-Net was demonstrated to use relevant regions of the optic chiasm for albinism detection from magnetic resonance imaging (MRI) brain anatomies. This indicates the strong potential of CNN-based approaches for visual pathway analysis and ultimately diagnostics.

Humeral shaft fractures: a practical guide to assessment and management.

Fractures of the humeral shaft represent roughly 5% of all fractures. They occur in an approximate bimodal distribution, typically affecting young adults following trauma and older females after low energy falls in the presence of osteoporosis. Humeral shaft fractures are associated with pain, temporary disability and a reduced quality of life for the duration of treatment. Treatment goals are directed towards achieving and maintaining a fracture environment conducive to healing, pain relief and early restoration of function. While most humeral shaft fractures are conservatively managed, operative management is indicated in certain circumstances. This article provides an overview of these fractures, including their initial management approach and definitive treatment.

Factors predicting the transition from acute to persistent pain in people with 'sciatica': the FORECAST longitudinal prognostic factor cohort study protocol.

INTRODUCTION: Sciatica is a common condition and is associated with higher levels of pain, disability, poorer quality of life, and increased use of health resources compared with low back pain alone. Although many patients recover, a third develop persistent sciatica symptoms. It remains unclear, why some patients develop persistent sciatica as none of the traditionally considered clinical parameters (eg, symptom severity, routine MRI) are consistent prognostic factors.The FORECAST study (factors predicting the transition from acute to persistent pain in people with 'sciatica') will take a different approach by exploring mechanism-based subgroups in patients with sciatica and investigate whether a mechanism-based approach can identify factors that predict pain persistence in patients with sciatica. METHODS AND ANALYSIS: We will perform a prospective longitudinal cohort study including 180 people with acute/subacute sciatica. N=168 healthy participants will provide normative data. A detailed set of variables will be assessed within 3 months after sciatica onset. This will include self-reported sensory and psychosocial profiles, quantitative sensory testing, blood inflammatory markers and advanced neuroimaging. We will determine outcome with the Sciatica Bothersomeness Index and a Numerical Pain Rating Scale for leg pain severity at 3 and 12 months.We will use principal component analysis followed by clustering methods to identify subgroups. Univariate associations and machine learning methods optimised for high dimensional small data sets will be used to identify the most powerful predictors and model selection/accuracy.The results will provide crucial information about the pathophysiological drivers of sciatica symptoms and may identify prognostic factors of pain persistence. ETHICS AND DISSEMINATION: The FORECAST study has received ethical approval (South Central Oxford C, 18/SC/0263). The dissemination strategy will be guided by our patient and public engagement activities and will include peer-reviewed publications, conference presentations, social media and podcasts. TRIAL REGISTRATION NUMBER: ISRCTN18170726; Pre-results.

Developing an understanding of artificial intelligence lung nodule risk prediction using insights from the Brock model.

OBJECTIVES: To determine if predictions of the Lung Cancer Prediction convolutional neural network (LCP-CNN) artificial intelligence (AI) model are analogous to the Brock model. METHODS: In total, 10,485 lung nodules in 4660 participants from the National Lung Screening Trial (NLST) were analysed. Both manual and automated nodule measurements were inputted into the Brock model, and this was compared to LCP-CNN. The performance of an experimental AI model was tested after ablating imaging features in a manner analogous to removing predictors from the Brock model. First, the nodule was ablated leaving lung parenchyma only. Second, a sphere of the same size as the nodule was implanted in the parenchyma. Third, internal texture of both nodule and parenchyma was ablated. RESULTS: Automated axial diameter (AUC 0.883) and automated equivalent spherical diameter (AUC 0.896) significantly improved the accuracy of Brock when compared to manual measurement (AUC 0.873), although not to the level of the LCP-CNN (AUC 0.936). Ablating nodule and parenchyma texture (AUC 0.915) led to a small drop in AI predictive accuracy, as did implanting a sphere of the same size as the nodule (AUC 0.889). Ablating the nodule leaving parenchyma only led to a large drop in AI performance (AUC 0.717). CONCLUSIONS: Feature ablation is a feasible technique for understanding AI model predictions. Nodule size and morphology play the largest role in AI prediction, with nodule internal texture and background parenchyma playing a limited role. This is broadly analogous to the relative importance of morphological factors over clinical factors within the Brock model. KEY POINTS: • Brock lung cancer risk prediction accuracy was significantly improved using automated axial or equivalent spherical measurements of lung nodule diameter, when compared to manual measurements. • Predictive accuracy was further improved by using the Lung Cancer Prediction convolutional neural network, an artificial intelligence-based model which obviates the requirement for nodule measurement. • Nodule size and morphology are important factors in artificial intelligence lung cancer risk prediction, with nodule texture and background parenchyma contributing a small, but measurable, role.

Lung cancer prediction by Deep Learning to identify benign lung nodules.

INTRODUCTION: Deep Learning has been proposed as promising tool to classify malignant nodules. Our aim was to retrospectively validate our Lung Cancer Prediction Convolutional Neural Network (LCP-CNN), which was trained on US screening data, on an independent dataset of indeterminate nodules in an European multicentre trial, to rule out benign nodules maintaining a high lung cancer sensitivity. METHODS: The LCP-CNN has been trained to generate a malignancy score for each nodule using CT data from the U.S. National Lung Screening Trial (NLST), and validated on CT scans containing 2106 nodules (205 lung cancers) detected in patients from from the Early Lung Cancer Diagnosis Using Artificial Intelligence and Big Data (LUCINDA) study, recruited from three tertiary referral centers in the UK, Germany and Netherlands. We pre-defined a benign nodule rule-out test, to identify benign nodules whilst maintaining a high sensitivity, by calculating thresholds on the malignancy score that achieve at least 99 % sensitivity on the NLST data. Overall performance per validation site was evaluated using Area-Under-the-ROC-Curve analysis (AUC). RESULTS: The overall AUC across the European centers was 94.5 % (95 %CI 92.6-96.1). With a high sensitivity of 99.0 %, malignancy could be ruled out in 22.1 % of the nodules, enabling 18.5 % of the patients to avoid follow-up scans. The two false-negative results both represented small typical carcinoids. CONCLUSION: The LCP-CNN, trained on participants with lung nodules from the US NLST dataset, showed excellent performance on identification of benign lung nodules in a multi-center external dataset, ruling out malignancy with high accuracy in about one fifth of the patients with 5-15 mm nodules.

Pelvic MRI in spinal cord injury patients: incidence of muscle signal change and early heterotopic ossification.

STUDY DESIGN: Prospective observational study. OBJECTIVE: To evaluate pelvic MRI muscle signal changes and their association with early heterotopic ossification (HO) in patients with spinal cord injuries. SETTING: National Spinal Injuries Unit, Stoke Mandeville, UK. METHODS: Forty patients were imaged with at least two interval magnetic resonance (MR) studies of the pelvis in the first 6 months following a spinal cord injury. Scans were reviewed and scored for heterotopic ossification, muscle signal change and extent of muscle involvement. RESULTS: Muscle signal change was present in 28 (70%) on the initial MRI and 31 (77%) by the second study. Six patients developed MR changes of prodromal or immature heterotopic ossification (15%). No restricted diffusion was demonstrated and no patient developed mature HO. Patients developing MR changes of early HO were more likely to have grade 3 muscle changes. CONCLUSION: Increased T2 muscle signal is common following cord injury, is frequently progressive in the subacute period and is associated with complete injury and early MR signs of heterotopic ossification.

Dual-energy CT in the diagnosis of occult acute scaphoid injury: a direct comparison with MRI.

OBJECTIVES: Scaphoid injuries occult on plain radiography often require further imaging for definitive diagnosis. We investigate the utility of dual-energy computed tomography (DECT) for the detection of acute bone marrow oedema and fracture of scaphoid compared to MRI. MATERIALS AND METHODS: Twenty patients who presented acutely (without prior injury) to the emergency department with clinically suspected occult scaphoid fracture and had MRI of the wrist were prospectively recruited to have DECT (GE Revolution CT). Material decomposition images of the water-calcium base pair were generated and assessed in conjunction with the monochromatic images to permit correlation of marrow signal changes with any cortical disruption for fracture confirmation. The assessment was performed by two musculoskeletal radiologists blinded from MRI results. The statistical difference of MRI and reviewers' detection of acute bone oedema (1 = present, 0 = absent) was performed using the Friedman test (SPSS v.16). RESULTS: MRI showed acute scaphoid fracture and/or bone marrow oedema in 14/20 patients of which 6 also had cortical disruption. On DECT, reviewer A identified oedema in 13 and cortical disruption in 10 patients while reviewer B identified oedema in 10 and cortical disruption in seven of the 14 MRI positive patients. No statistically significant difference in oedema detection on MRI and reviewers of DECT (p value 0.61) but DECT was more sensitive at detecting cortical disruption. CONCLUSION: DECT has the capability to detect acute scaphoid oedema in addition to cortical fractures. However, compared to MRI, DECT has lower contrast resolution and less sensitive in the detection of mild oedema. KEY POINTS: • Dual-energy CT is able to detect acute traumatic scaphoid marrow oedema. • Dual-energy CT has greater detection rate of scaphoid fractures than MRI. • Dual-energy CT is an alternative to MRI for occult scaphoid injury.

Diagnostic accuracy of initial chest radiograph compared to SARS-CoV-2 PCR in patients with suspected COVID-19.

OBJECTIVE: The chest radiograph (CXR) is the predominant imaging investigation being used to triage patients prior to either performing a SARS-CoV-2 polymerase chain reaction (PCR) test or a diagnostic CT scan, but there are limited studies that assess the diagnostic accuracy of CXRs in COVID-19.To determine the accuracy of CXR diagnosis of COVID-19 compared with PCR in patients presenting with a clinical suspicion of COVID-19. METHODS AND MATERIALS: The CXR reports of 569 consecutive patients with a clinical suspicion of COVID-19 were reviewed, blinded to the PCR result and classified into the following categories: normal, indeterminate for COVID-19, classic/probable COVID-19, non-COVID-19 pathology, and not specified. Severity reporting and reporter expertise were documented. The subset of this cohort that had CXR and PCR within 3 days of each other were included for further analysis for diagnostic accuracy. RESULTS: Classic/probable COVID-19 was reported in 29% (166/569) of the initial cohort. 67% (382/569) had PCR tests. 344 patients had CXR and PCR within 3 days of each other. Compared to PCR as the reference test, initial CXR had a 61% sensitivity and 76% specificity in the diagnosis of COVID-19. CONCLUSION: Initial CXR is useful as a triage tool with a sensitivity of 61% and specificity of 76% in the diagnosis of COVID-19 in a hospital setting. ADVANCES IN KNOWLEDGE: .Diagnostic accuracy does not differ significantly between specialist thoracic radiologists and general radiologists including trainees following training.There was a 40% prevalence of PCR positive disease in the cohort of patients (n = 344) having CXR and PCR within 3 days of each other.Classic/probable COVID-19 was reported in 29% of total cohort of patients presenting with clinical suspicion of COVID-19 (n = 569).Initial CXR is useful as a triage tool with a sensitivity of 61% and specificity of 76% in the diagnosis of COVID-19 in a hospital setting.

Snapping Shoulder Secondary to Subluxation of an Accessory Coracobrachialis.

Numerous snapping syndromes have been reported in the musculoskeletal system. Identifying the cause of these symptoms can often be challenging as the underlying abnormality may not be appreciable on routine static examinations. We report a 30-year-old female who presented with an unusual snapping sensation in her right anterior shoulder. This was readily reproducible during shoulder abduction with a palpable clicking evident on clinical examination. Dynamic ultrasound revealed this to be secondary to an accessory coracobrachialis muscle, which subluxed suddenly over the anterior subscapularis tendon during abduction. An accessory coracobrachialis muscle is a rare normal variant that is often asymptomatic. Extra-articular causes of shoulder snapping have been rarely reported, and this is the first case report of an accessory coracobrachialis muscle causing a snapping shoulder phenomenon.

Assessing the Accuracy of a Deep Learning Method to Risk Stratify Indeterminate Pulmonary Nodules.

Rationale: The management of indeterminate pulmonary nodules (IPNs) remains challenging, resulting in invasive procedures and delays in diagnosis and treatment. Strategies to decrease the rate of unnecessary invasive procedures and optimize surveillance regimens are needed.Objectives: To develop and validate a deep learning method to improve the management of IPNs.Methods: A Lung Cancer Prediction Convolutional Neural Network model was trained using computed tomography images of IPNs from the National Lung Screening Trial, internally validated, and externally tested on cohorts from two academic institutions.Measurements and Main Results: The areas under the receiver operating characteristic curve in the external validation cohorts were 83.5% (95% confidence interval [CI], 75.4-90.7%) and 91.9% (95% CI, 88.7-94.7%), compared with 78.1% (95% CI, 68.7-86.4%) and 81.9 (95% CI, 76.1-87.1%), respectively, for a commonly used clinical risk model for incidental nodules. Using 5% and 65% malignancy thresholds defining low- and high-risk categories, the overall net reclassifications in the validation cohorts for cancers and benign nodules compared with the Mayo model were 0.34 (Vanderbilt) and 0.30 (Oxford) as a rule-in test, and 0.33 (Vanderbilt) and 0.58 (Oxford) as a rule-out test. Compared with traditional risk prediction models, the Lung Cancer Prediction Convolutional Neural Network was associated with improved accuracy in predicting the likelihood of disease at each threshold of management and in our external validation cohorts.Conclusions: This study demonstrates that this deep learning algorithm can correctly reclassify IPNs into low- or high-risk categories in more than a third of cancers and benign nodules when compared with conventional risk models, potentially reducing the number of unnecessary invasive procedures and delays in diagnosis.

Aberrant visual pathway development in albinism: From retina to cortex.

Albinism refers to a group of genetic abnormalities in melanogenesis that are associated neuronal misrouting through the optic chiasm. We perform quantitative assessment of visual pathway structure and function in 23 persons with albinism (PWA) and 20 matched controls using optical coherence tomography (OCT), volumetric magnetic resonance imaging (MRI), diffusion tensor imaging and visual evoked potentials (VEP). PWA had a higher streamline decussation index (percentage of total tractography streamlines decussating at the chiasm) compared with controls (Z = -2.24, p = .025), and streamline decussation index correlated weakly with inter-hemispheric asymmetry measured using VEP (r = .484, p = .042). For PWA, a significant correlation was found between foveal development index and total number of streamlines (r = .662, p < .001). Significant positive correlations were found between peri-papillary retinal nerve fibre layer thickness and optic nerve (r = .642, p < .001) and tract (r = .663, p < .001) width. Occipital pole cortical thickness was 6.88% higher (Z = -4.10, p < .001) in PWA and was related to anterior visual pathway structures including foveal retinal pigment epithelium complex thickness (r = -.579, p = .005), optic disc (r = .478, p = .021) and rim areas (r = .597, p = .003). We were unable to demonstrate a significant relationship between OCT-derived foveal or optic nerve measures and MRI-derived chiasm size or streamline decussation index. Our novel tractographic demonstration of altered chiasmatic decussation in PWA corresponds to VEP measured cortical asymmetry and is consistent with chiasmatic misrouting in albinism. We also demonstrate a significant relationship between retinal pigment epithelium and visual cortex thickness indicating that retinal pigmentation defects in albinism lead to downstream structural reorganisation of the visual cortex.

Development and validation of clinical prediction models to risk stratify patients presenting with small pulmonary nodules: a research protocol.

INTRODUCTION: Lung cancer is a common cancer, with over 1.3 million cases worldwide each year. Early diagnosis using computed tomography (CT) screening has been shown to reduce mortality but also detect non-malignant nodules that require follow-up scanning or alternative methods of investigation. Practical and accurate tools that can predict the probability that a lung nodule is benign or malignant will help reduce costs and the risk of morbidity and mortality associated with lung cancer. METHODS: Retrospectively collected data from 1500 patients with pulmonary nodule(s) of up to 15 mm detected on routinely performed CT chest scans aged 18 years old or older from three academic centres in the UK will be used to to develop risk stratification models. Radiological, clinical and patient characteristics will be combined in multivariable logistic regression models to predict nodule malignancy. Data from over 1000 participants recruited in a prospective phase of the study will be used to evaluate model performance. Discrimination, calibration and clinical utility measures will be presented.