Automatic Detection and Classification of Multiple Catheters in Neonatal Radiographs with Deep Learning

We develop and evaluate a deep learning algorithm to classify multiple catheters on neonatal chest and abdominal radiographs. A convolutional neural network (CNN) was trained using a dataset of 777 neonatal chest and abdominal radiographs, with a split of 81%-9%-10% for training-validation-testing, respectively. We employed ResNet-50 (a CNN), pre-trained on ImageNet. Ground truth labelling was limited to tagging each image to indicate the presence or absence of endotracheal tubes (ETTs), nasogastric tubes (NGTs), and umbilical arterial and venous catheters (UACs, UVCs). The dataset included 561 images containing two or more catheters, 167 images with only one, and 49 with none. Performance was measured with average precision (AP), calculated from the area under the precision-recall curve. On our test data, the algorithm achieved an overall AP (95% confidence interval) of 0.977 (0.679-0.999) for NGTs, 0.989 (0.751-1.000) for ETTs, 0.979 (0.873-0.997) for UACs, and 0.937 (0.785-0.984) for UVCs. Performance was similar for the set of 58 test images consisting of two or more catheters, with an AP of 0.975 (0.255-1.000) for NGTs, 0.997 (0.009-1.000) for ETTs, 0.981 (0.797-0.998) for UACs, and 0.937 (0.689-0.990) for UVCs. Our network thus achieves strong performance in the simultaneous detection of these four catheter types. Radiologists may use such an algorithm as a time-saving mechanism to automate reporting of catheters on radiographs.

Development and Cost Analysis of a Lung Nodule Management Strategy Combining Artificial Intelligence and Lung-RADS for Baseline Lung Cancer Screening

OBJECTIVES

To develop a lung nodule management strategy combining the Lung CT Screening Reporting and Data System (Lung-RADS) with an artificial intelligence (AI) malignancy risk score and determine its impact on follow-up investigations and associated costs in a baseline lung cancer screening population.

MATERIALS AND METHODS

Secondary analysis was undertaken of a data set consisting of AI malignancy risk scores and Lung-RADS classifications from six radiologists for 192 baseline low-dose CT studies. Low-dose CT studies were weighted to model a representative cohort of 3,197 baseline screening patients. An AI risk score threshold was defined to match average sensitivity of six radiologists applying Lung-RADS. Cases initially Lung-RADS category 1 or 2 with a high AI risk score were upgraded to category 3, and cases initially category 3 or higher with a low AI risk score were downgraded to category 2. Follow-up investigations resulting from Lung-RADS and the AI-informed management strategy were determined. Investigation costs were based on the 2019 US Medicare Physician Fee Schedule.

RESULTS

The AI-informed management strategy achieved sensitivity and specificity of 91% and 96%, respectively. Average sensitivity and specificity of six radiologists using Lung-RADS only was 91% and 66%, respectively. Using the AI-informed management strategy, 41 (0.2%) category 1 or 2 classifications were upgraded to category 3, and 5,750 (30%) category 3 or higher classifications were downgraded to category 2. Minimum net cost savings using the AI-informed management strategy was estimated to be $72 per patient screened.

CONCLUSION

Using an AI risk score combined with Lung-RADS at baseline lung cancer screening may result in fewer follow-up investigations and substantial cost savings.

Artificial intelligence-based imaging analytics and lung cancer diagnostics: Considerations for health system leaders

Lung cancer is a leading cause of cancer death in Canada, and accurate, early diagnosis are critical to improving clinical outcomes. Artificial Intelligence (AI)-based imaging analytics are a promising healthcare innovation that aim to improve the accuracy and efficiency of lung cancer diagnosis. Maximizing their clinical potential while mitigating their risks and limitations will require focused leadership informed by interdisciplinary expertise and system-wide insight. We convened a knowledge exchange workshop with diverse Saskatchewan health system leaders and stakeholders to explore issues surrounding the use of AI in diagnostic imaging for lung cancer, including implementation opportunities, challenges, and priorities. This technology is anticipated to improve patient outcomes, reduce unnecessary healthcare spending, and increase knowledge. However, health system leaders must also address the needs for robust data, financial investment, effective communication and collaboration between healthcare sectors, privacy and data protections, and continued interdisciplinary research to achieve this technology's potential benefits.

Telerobotic ultrasound to provide obstetrical ultrasound services remotely during the COVID-19 pandemic

Introduction

Obstetrical ultrasound imaging is critical in identifying at-risk pregnancies and informing clinical management. The coronavirus disease 2019 (COVID-19) pandemic has exacerbated challenges in accessing obstetrical ultrasound for patients in underserved rural and remote communities where this service is not available. This prospective descriptive study describes our experience of providing obstetrical ultrasound services remotely using a telerobotic ultrasound system in a northern Canadian community isolated due to a COVID-19 outbreak.

Methods

A telerobotic ultrasound system was used to perform obstetrical ultrasound exams remotely in La Loche, Canada, a remote community without regular access to obstetrical ultrasound. Using a telerobotic ultrasound system, a sonographer 605 km away remotely controlled an ultrasound probe and ultrasound settings. Twenty-one exams were performed in a five-week period during a COVID-19 outbreak in the community, including limited first-, second- and third-trimester exams (n = 11) and complete second-trimester exams (n = 10). Participants were invited to complete a survey at the end of the telerobotic ultrasound exam describing their experiences with telerobotic ultrasound. Radiologists subsequently interpreted all exams and determined the adequacy of the images for diagnosis.

Results

Of 11 limited obstetrical exams, radiologists indicated images were adequate in nine (81%) cases, adequate with some reservations in one (9%) case and inadequate in one (9%) case. Of 10 second-trimester complete obstetrical exams, radiologists indicated images were adequate in two (20%) cases, adequate with some reservations in three (30%) cases and inadequate in five (50%) cases. Second-trimester complete obstetrical exams were limited due to a combination of body habitus, foetal lie and telerobotic technology.

Discussion

A telerobotic ultrasound system may be used to answer focused clinical questions such as foetal viability, dating and foetal presentation in a timely manner while minimising patient travel to larger centres and potential exposure to severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), during the COVID-19 pandemic.

Artificial Intelligence Solutions for Analysis of X-ray Images

Artificial intelligence (AI) presents a key opportunity for radiologists to improve quality of care and enhance the value of radiology in patient care and population health. The potential opportunity of AI to aid in triage and interpretation of conventional radiographs (X-ray images) is particularly significant, as radiographs are the most common imaging examinations performed in most radiology departments. Substantial progress has been made in the past few years in the development of AI algorithms for analysis of chest and musculoskeletal (MSK) radiographs, with deep learning now the dominant approach for image analysis. Large public and proprietary image data sets have been compiled and have aided the development of AI algorithms for analysis of radiographs, many of which demonstrate accuracy equivalent to radiologists for specific, focused tasks. This article describes (1) the basis for the development of AI solutions for radiograph analysis, (2) current AI solutions to aid in the triage and interpretation of chest radiographs and MSK radiographs, (3) opportunities for AI to aid in noninterpretive tasks related to radiographs, and (4) considerations for radiology practices selecting AI solutions for radiograph analysis and integrating them into existing IT systems. Although comprehensive AI solutions across modalities have yet to be developed, institutions can begin to select and integrate focused solutions which increase efficiency, increase quality and patient safety, and add value for their patients.

Computer-aided Assessment of Catheters and Tubes on Radiographs: How Good Is Artificial Intelligence for Assessment?

Catheters are the second most common abnormal finding on radiographs. The position of catheters must be assessed on all radiographs because serious complications can arise if catheters are malpositioned. However, due to the large number of radiographs obtained each day, there can be substantial delays between the time a radiograph is obtained and when it is interpreted by a radiologist. Computer-aided approaches hold the potential to assist in prioritizing radiographs with potentially malpositioned catheters for interpretation and automatically insert text indicating the placement of catheters in radiology reports, thereby improving radiologists' efficiency. After 50 years of research in computer-aided diagnosis, there is still a paucity of study in this area. With the development of deep learning approaches, the problem of catheter assessment is far more solvable. This review provides an overview of current algorithms and identifies key challenges in building a reliable computer-aided diagnosis system for assessment of catheters on radiographs. This review may serve to further the development of machine learning approaches for this important use case. Supplemental material is available for this article. © RSNA, 2020.

Development of a risk assessment model to differentiate malignant and benign musculoskeletal soft-tissue masses on magnetic resonance imaging

INTRODUCTION

This study aimed to develop a risk stratification model to differentiate benign and malignant MRI-imaged musculoskeletal soft-tissue tumours, informing decisions surrounding biopsy and follow-up imaging.

METHODS

Imaging of patients who underwent MRI and subsequent biopsy to evaluate a soft-tissue mass was retrospectively reviewed. Features analysed included patient age; tumour size; shape; margins; enhancement pattern; signal intensity pattern; deep fascia, neurovascular bundle, bone and joint involvement; and the presence of necrosis, haemorrhage, oedema and intralesional fat. Univariate comparisons, by final histopathological status, employed t-tests and chi-square tests, followed by simple and multiple logistic regressions. Variables included in the final multiple regression model were used to define a three-level risk stratification strategy.

RESULTS

One-hundred and ten patients were included in the analysis. Univariate relationships were identified between malignancy and age, tumour size, deep fascia involvement, neurovascular involvement, necrosis, haemorrhage, oedema and heterogeneous enhancement (all P < 0.01). Final multiple regression modelling included size, enhancement and oedema. Thirty of 40 (75%) tumours >5 cm with surrounding oedema ('high risk') were malignant, 13 of 47 (28%) tumours with one or more of tumour size >5 cm, surrounding oedema or heterogeneous enhancement ('moderate risk') were malignant, and none of the 16 tumours ≤5 cm with the absence of surrounding oedema and heterogeneous enhancement ('low risk') were malignant.

CONCLUSIONS

A model including tumour size, enhancement and oedema has potential to stratify soft-tissue tumours into high-, intermediate- and low-risk categories; this may inform decisions surrounding biopsy and follow-up imaging.

A Crossover Comparison of Standard and Telerobotic Approaches to Prenatal Sonography

OBJECTIVES

To determine the feasibility of a telerobotic approach to remotely perform prenatal sonographic examinations.

METHODS

Thirty participants were prospectively recruited. Participants underwent a limited examination (assessing biometry, placental location, and amniotic fluid; n = 20) or a detailed examination (biometry, placental location, amniotic fluid, and fetal anatomic survey; n = 10) performed with a conventional ultrasound system. This examination was followed by an equivalent examination performed with a telerobotic ultrasound system, which enabled sonographers to remotely control all ultrasound settings and fine movements of the ultrasound transducer from a distance. Telerobotic images were read independently from conventional images.

RESULTS

The mean gestational age ± SD of the 30 participants was 22.9 ± 5.3 weeks. Paired-sample t tests showed no statistically significant difference between conventional and telerobotic measurements of fetal head circumference, biparietal diameter, or single deepest vertical pocket of amniotic fluid; however, a small but statistically significant difference was observed in measurements of abdominal circumference and femur length (P < .05). Intraclass correlations showed excellent agreement (>0.90) between telerobotic and conventional measurements of all 4 biometric parameters. Of 21 fetal structures included in the anatomic survey, 80% of the structures attempted across all patients were sufficiently visualized by the telerobotic system (range, 57%-100% per patient). Ninety-seven percent of patients strongly or somewhat agreed that they would be willing to have another telerobotic examination in the future.

CONCLUSIONS

A telerobotic approach is feasible for remotely performing prenatal sonographic examinations. Telerobotic sonography (robotic telesonography) may allow for the development of satellite ultrasound clinics in rural, remote, or low-volume communities, thereby increasing access to prenatal imaging in underserved communities.

Adult-onset leukoencephalopathy with axonal spheroids and pigmented glia (ALSP): Integrating the literature on hereditary diffuse leukoencephalopathy with spheroids (HDLS) and pigmentary orthochromatic leukodystrophy (POLD)

Adult-onset leukoencephalopathy with axonal spheroids and pigmented glia (ALSP) is a progressive degenerative white matter disorder. ALSP was previously recognized as two distinct entities, hereditary diffuse leukoencephalopathy with spheroids (HDLS) and pigmentary orthochromatic leukodystrophy (POLD). However, recent identification of mutations in the tyrosine kinase domain of the colony stimulating factor 1 receptor (CSF1R) gene, which regulates mononuclear cell lineages including microglia, have provided genetic and mechanistic evidence that POLD and HDLS should be regarded as a single clinicopathologic entity. We describe two illustrative cases of ALSP which presented with neuropsychiatric symptoms, progressive cognitive decline, and motor and gait disturbances. Antemortem diagnoses of autopsy-confirmed ALSP vary significantly, and include primary progressive multiple sclerosis, frontotemporal dementia, Alzheimer disease, atypical cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL), corticobasal syndrome, and atypical Parkinson disease, suggesting that ALSP may be significantly underdiagnosed. This article presents a systematic review of ALSP in the context of two illustrative cases to help integrate the literature on HDLS and POLD. Consistent use of the term ALSP is suggested for clarity in the literature going forward.

Invasive lobular carcinoma of the male breast - a systematic review with an illustrative case study

Male breast cancer is rare, comprising only 1% of all mammary cancers; invasive ductal carcinoma is by far the commonest subtype in both men and women. Though lobular breast cancer is the second most common subtype seen in women, such cancers are extremely uncommon in men, and this is likely related to the lack of lobular development in the male breast. Thus, due to the rarity of this subtype among breast cancers, compounded by the overall rarity of breast cancer in men, current understanding of the pathogenesis of this disease and its management is largely derived from case series and extrapolation of information from the larger cohort of female patients. This paper provides a systematic review on invasive lobular carcinoma of the male breast in the context of an illustrative case study. A comprehensive analysis of the National Cancer Institute's Surveillance, Epidemiology, and End Results Data 1973-2013 leading to an exploration of the pathogenesis, epidemiology, clinical presentation, diagnosis, tumor characteristics, and management of lobular breast carcinoma in men is also discussed. Lobular subtype of breast cancer remains an enigmatic elusive disease that needs additional research to unravel its overall pathogenesis and molecular profile to provide insight for improved therapeutic management options.

Initial Experience Using a Telerobotic Ultrasound System for Adult Abdominal Sonography

PURPOSE

The study sought to assess the feasibility of performing adult abdominal examinations using a telerobotic ultrasound system in which radiologists or sonographers can control fine movements of a transducer and all ultrasound settings from a remote location.

METHODS

Eighteen patients prospectively underwent a conventional sonography examination (using EPIQ 5 [Philips] or LOGIQ E9 [GE Healthcare]) followed by a telerobotic sonography examination (using the MELODY System [AdEchoTech] and SonixTablet [BK Ultrasound]) according to a standardized abdominal imaging protocol. For telerobotic examinations, patients were scanned remotely by a sonographer 2.75 km away. Conventional examinations were read independently from telerobotic examinations. Image quality and acceptability to patients and sonographers was assessed.

RESULTS

Ninety-two percent of organs visualized on conventional examinations were sufficiently visualized on telerobotic examinations. Five pathological findings were identified on both telerobotic and conventional examinations, 3 findings were identified using only conventional sonography, and 2 findings were identified using only telerobotic sonography. A paired sample t test showed no significant difference between the 2 modalities in measurements of the liver, spleen, and diameter of the proximal aorta; however, telerobotic assessments overestimated distal aorta and common bile duct diameters and underestimated kidney lengths (P values < .05). All patients responded that they would be willing to have another telerobotic examination.

CONCLUSIONS

A telerobotic ultrasound system is feasible for performing abdominal ultrasound examinations at a distant location with minimal training and setup requirements and a moderate learning curve. Telerobotic sonography (robotic telesonography) may open up the possibility of remote ultrasound clinics for communities that lack skilled sonographers and radiologists, thereby improving access to care.

Cost analysis of a home-based nurse care coordination program

Objectives

To determine whether a home-based care coordination program focused on medication self-management would affect the cost of care to the Medicare program and whether the addition of technology, a medication-dispensing machine, would further reduce cost.

Design

Randomized, controlled, three-arm longitudinal study.

Setting

Participant homes in a large Midwestern urban area.

Participants

Older adults identified as having difficulty managing their medications at discharge from Medicare Home Health Care (N = 414).

Intervention

A team consisting of advanced practice nurses (APNs) and registered nurses (RNs) coordinated care for two groups: home-based nurse care coordination (NCC) plus a pill organizer group and NCC plus a medication-dispensing machine group.

Measurements

To measure cost, participant claims data from 2005 to 2011 were retrieved from Medicare Part A and B Standard Analytical Files.

Results

Ordinary least squares regression with covariate adjustment was used to estimate monthly dollar savings. Total Medicare costs were $447 per month lower in the NCC plus pill organizer group (P = .11) than in a control group that received usual care. For participants in the study at least 3 months, total Medicare costs were $491 lower per month in the NCC plus pill organizer group (P = .06) than in the control group. The cost of the NCC plus pill organizer intervention was $151 per month, yielding a net savings of $296 per month or $3,552 per year. The cost of the NCC plus medication-dispensing machine intervention was $251 per month, and total Medicare costs were $409 higher per month than in the NCC plus pill organizer group.

Conclusion

Nurse care coordination plus a pill organizer is a cost-effective intervention for frail elderly Medicare beneficiaries. The addition of the medication machine did not enhance the cost effectiveness of the intervention.

Use of the Tecpro Pork Quality Meter for assessing meat quality on the slaughterline

Conductivity measurements made with the Tecpro Pork Quality Meter at either 45 min or about 20 h post mortem on 224 pig cacassees which exhibited a wide range of raw meat quality were moderately correlated with initial pH (pH(45); r = 0.54 and 0.62) but less well correlated with reflectance (r = 0.32 and 0.37), drip loss (r = 0.34 and 0.47 and other objective measures of meat quality in the M. Longissimus dorsi. Measurements did not allow differentiation between normal and dark, firm, dry (DFD) meat but were of some value in identifying carcasses producing potentially pole, soft, exudative (PSE) meat. In the UK they would be more useful for monitoring meat quality in large groups or populations of pigs, than for categorising individual carcasses for grading on technological purposes.

Aging in place versus nursing home care: comparison of costs to Medicare and Medicaid

The objective of this study was to compare the community-based, long-term care program called Aging in Place (AIP) and nursing home care, in terms of cost to the Medicare and Medicaid programs. A retrospective cohort design was used in this study of 39 nursing home residents in the Midwest who were matched with 39 AIP participants. The AIP program consisted of a combination of Medicare home health, Medicaid home and community-based services (HCBS), and intensive nurse care coordination. Controlling for high inpatient Medicare cost in the 6 months prior and the 10 most frequently occurring chronic conditions, multiple regression was used to estimate the relationship of the AIP program on Medicare and Medicaid costs. Total Medicare and Medicaid costs were $1,591.61 lower per month in the AIP group (p < 0.01) when compared with the nursing home group over a 12-month period. The findings suggest that the provision of nurse-coordinated HCBS and Medicare home health services has potential to provide savings in the total cost of health care to the Medicaid program while not increasing the cost of the Medicare program.

The relationship of community-based nurse care coordination to costs in the Medicare and Medicaid programs

The purpose of this evaluation was to study the relationship of nurse care coordination (NCC) to the costs of Medicare and Medicaid in a community-based care program called Missouri Care Options (MCO). A retrospective cohort design was used comparing 57 MCO clients with NCC to 80 MCO clients without NCC. Total cost was measured using Medicare and Medicaid claims databases. Fixed effects analysis was used to estimate the relationship of the NCC intervention to costs. Controlling for high resource use on admission, monthly Medicare costs were lower ($686) in the 12 months of NCC intervention (p = .04) while Medicaid costs were higher ($203; p = .03) for the NCC group when compared to the costs of MCO group.

GATA4 is essential for jejunal function in mice

BACKGROUND & AIMS

Although the zinc-finger transcription factor GATA4 has been implicated in regulating jejunal gene expression, the contribution of GATA4 in controlling jejunal physiology has not been addressed.

METHODS

We generated mice in which the Gata4 gene was specifically deleted in the small intestinal epithelium. Measurements of plasma cholesterol and phospholipids, intestinal absorption of dietary fat and cholesterol, and gene expression were performed on these animals.

RESULTS

Mice lacking GATA4 in the intestine displayed a dramatic block in their ability to absorb cholesterol and dietary fat. Comparison of the global gene expression profiles of control jejunum, control ileum, and GATA4 null jejunum by gene array analysis revealed that GATA4 null jejunum lost expression of 53% of the jejunal-specific gene set and gained expression of 47% of the set of genes unique to the ileum. These alterations in gene expression included a decrease in messenger RNAs (mRNAs) encoding lipid and cholesterol transporters as well as an increase in mRNAs encoding proteins involved in bile acid absorption.

CONCLUSIONS

Our data demonstrate that GATA4 is essential for jejunal function including fat and cholesterol absorption and confirm that GATA4 plays a pivotal role in determining jejunal vs ileal identity.

Illness perception and health-related quality of life in multiple sclerosis

AIMS - A number of physical and psychological factors have been shown to affect health-related quality of life (HRQoL) in patients with multiple sclerosis (MS). Among these, the role of illness perceptions has not been established as an independent factor. This study, the first of its kind in an Australian population, aimed to use a large sample to determine the relative importance of individual factors to each domain of HRQoL, in particular the role of illness perception. MATERIALS AND METHODS - 580 patients with confirmed MS were assessed cross sectionally in a designated research clinic to determine the relative impact of physical factors (illness severity, duration, age, fatigue and pain) and psychological factors (mood, cognition and illness representations) on each domain of the SF-36. RESULTS - Categorical regression analysis showed that a combination of physical and psychological factors predicted 38-71% of variance in HRQoL. Illness perception was shown to have an independent effect on HRQoL in MS. The Extended Disability Status Scale was a significant determinant in all domains except for mental health. Depression was less prevalent than anxiety, but had a greater effect on function. CONCLUSION - Illness perception is an independent factor contributing to HRQoL in people with MS. Individual domains of HRQoL are associated with different patterns of physical and psychological factors. In the domains of role and social function, activities most highly valued by patients with MS, depression, anxiety, fatigue and illness perceptions are key determinants, all of which have the potential to be improved through specific interventions.

Differential effectiveness of depression disease management for rural and urban primary care patients

CONTEXT

Federally qualified health centers across the country are adopting depression disease management programs following federally mandated training; however, little is known about the relative effectiveness of depression disease management in rural versus urban patient populations.

PURPOSE

To explore whether a depression disease management program has a comparable impact on clinical outcomes over 2 years in patients treated in rural and urban primary care practices and whether the impact is mediated by receiving evidence-based care (antidepressant medication and specialty care counseling).

METHODS

A preplanned secondary analysis was conducted in a consecutively sampled cohort of 479 depressed primary care patients recruited from 12 practices in 10 states across the country participating in the Quality Enhancement for Strategic Teaming study.

FINDINGS

Depression disease management improved the mental health status of urban patients over 18 months but not rural patients. Effects were not mediated by antidepressant medication or specialty care counseling in urban or rural patients.

CONCLUSIONS

Depression disease management appears to improve clinical outcomes in urban but not rural patients. Because these programs compete for scarce resources, health care organizations interested in delivering depression disease management to rural populations need to advocate for programs whose clinical effectiveness has been demonstrated for rural residents.