Economic Evaluations and Equity in the Use of Artificial Intelligence in Imaging Exams for Medical Diagnosis in People With Skin, Neurological, and Pulmonary Diseases: Protocol for a Systematic Review

BACKGROUND

Traditional health care systems face long-standing challenges, including patient diversity, geographical disparities, and financial constraints. The emergence of artificial intelligence (AI) in health care offers solutions to these challenges. AI, a multidisciplinary field, enhances clinical decision-making. However, imbalanced AI models may enhance health disparities.

OBJECTIVE

This systematic review aims to investigate the economic performance and equity impact of AI in diagnostic imaging for skin, neurological, and pulmonary diseases. The research question is "To what extent does the use of AI in imaging exams for diagnosing skin, neurological, and pulmonary diseases result in improved economic outcomes, and does it promote equity in health care systems?"

METHODS

The study is a systematic review of economic and equity evaluations following PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) and CHEERS (Consolidated Health Economic Evaluation Reporting Standards) guidelines. Eligibility criteria include articles reporting on economic evaluations or equity considerations related to AI-based diagnostic imaging for specified diseases. Data will be collected from PubMed, Embase, Scopus, Web of Science, and reference lists. Data quality and transferability will be assessed according to CHEC (Consensus on Health Economic Criteria), EPHPP (Effective Public Health Practice Project), and Welte checklists.

RESULTS

This systematic review began in March 2023. The literature search identified 9,526 publications and, after full-text screening, 9 publications were included in the study. We plan to submit a manuscript to a peer-reviewed journal once it is finalized, with an expected completion date in January 2024.

CONCLUSIONS

AI in diagnostic imaging offers potential benefits but also raises concerns about equity and economic impact. Bias in algorithms and disparities in access may hinder equitable outcomes. Evaluating the economic viability of AI applications is essential for resource allocation and affordability. Policy makers and health care stakeholders can benefit from this review's insights to make informed decisions. Limitations, including study variability and publication bias, will be considered in the analysis. This systematic review will provide valuable insights into the economic and equity implications of AI in diagnostic imaging. It aims to inform evidence-based decision-making and contribute to more efficient and equitable health care systems.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

DERR1-10.2196/48544.

The Neurobiology of Zika Virus: New Models, New Challenges

The Zika virus (ZIKV) attracted attention due to one striking characteristic: the ability to cross the placental barrier and infect the fetus, possibly causing severe neurodevelopmental disruptions included in the Congenital Zika Syndrome (CZS). Few years after the epidemic, the CZS incidence has begun to decline. However, how ZIKV causes a diversity of outcomes is far from being understood. This is probably driven by a chain of complex events that relies on the interaction between ZIKV and environmental and physiological variables. In this review, we address open questions that might lead to an ill-defined diagnosis of CZS. This inaccuracy underestimates a large spectrum of apparent normocephalic cases that remain underdiagnosed, comprising several subtle brain abnormalities frequently masked by a normal head circumference. Therefore, new models using neuroimaging and artificial intelligence are needed to improve our understanding of the neurobiology of ZIKV and its true impact in neurodevelopment.

Sensorimotor white matter projections and disease severity in primary Restless Legs Syndrome/Willis-Ekbom disease: a multimodal DTI analysis

BACKGROUND

Restless Legs Syndrome, a potentially disabling sleep disorder, also known as Willis-Ekbom disease (RLS/WED), may be caused by loss of inhibitory modulation of descending central motor pathways, structural changes in the somatosensory cortex, abnormal connectivity between motor and sensory areas, as well as by subtle abnormalities in white matter micro-organization.

OBJECTIVE

To compare diffusion-tensor imaging (DTI) metrics in areas associated with sensory or motor function, as well as sensorimotor integration, between subjects with primary mild-to-severe RLS/WED and controls.

METHODS

DTI metrics were assessed in 38 subjects with RLS/WED (14 mild to moderate, 24 severe to very severe) and 24 healthy age-matched controls with whole-brain Tract Based Spatial Statistics (TBSS), Region-of-interest (ROI) and probabilistic tractography based analyses. The ROIs corresponded to the corticospinal tract (CST) at the level of the cerebral peduncle; the superior, middle and inferior cerebellar peduncles. Subgroup analyses were made according to the severity of RLS/WED symptoms. The corticospinal tract was evaluated with probabilistic tractography. We also explored associations between significant findings and severity of symptoms with the Spearman's correlation coefficient.

RESULTS

TBSS analysis revealed decreased axial diffusivity (AD) in the left posterior thalamic radiation in RLS/WED. In subjects with severe RLS/WED, AD was reduced in the left posterior corona radiata and this reduction was negatively correlated with severity of symptoms. ROI-based analysis showed that radial diffusivity (RD) was increased in the superior cerebellar peduncles of individuals with severe RLS/WED. Tractography did not show between-group or subgroup differences.

CONCLUSIONS

Our results are consistent with subtle white matter changes, prominently in RLS/WED subjects with more severe symptoms, in areas related to sensory or motor function, as well as to sensorimotor integration, compared to controls. These findings support the hypothesis, raised by prior pathophysiological studies, of defective integration within these networks.

Relationship between footshock intensity, post-training corticosterone release and contextual fear memory specificity over time

Overgeneralized fear has long been implicated in generalized anxiety and post-traumatic stress disorder, however, time-dependent mechanisms underlying memory retrieval are still not completely understood. Previous studies have revealed that stronger fear conditioning training protocols are associated with both increased post-training corticosterone (CORT) levels and fear responses at later retrieval tests. Here we used contextual fear conditioning (CFC) to investigate the relationship between post-training CORT levels and memory specificity in different retrieval timepoints. Wistar rats were exposed to CFC training with increasing footshock intensities (0.3, 0.6 or 1.0mA) and had their blood collected 30 min afterwards to measure post-training plasma CORT. After 2, 14 or 28 days, rats were tested for memory specificity either in the training or in the novel context. Regression analysis was used to verify linear and non-linear interactions between CORT levels and freezing. Higher footshock intensities increased post-training CORT levels and freezing times during tests in all timepoints. Moreover, stronger trainings elicited faster memory generalization, which was associated with higher CORT levels during memory consolidation. The 0.3mA training maintained memory specificity up to 28 days. Additionally, linear regressions suggest that the shift from specific to generalized memories is underway at 14 days after training. These results are consistent with the hypotheses that stronger training protocols elicit a faster generalization rate, and that this process is associated with increased post-training CORT release.

Short-interval intracortical inhibition is decreased in restless legs syndrome across a range of severity

BACKGROUND

Decreased short-interval intracortical inhibition (SICI) to transcranial magnetic stimulation (TMS) of the primary motor cortex was described in subjects with restless legs syndrome/Willis-Ekbom disease (RLS/WED). It remained to be determined whether the magnitude of SICI decrease would be similar across levels of RLS/WED severity. Moreover, it was unknown whether, in addition to decreases in SICI, changes in cortical thickness or area could be detected in subjects with RLS/WED compared to controls. The objective of this study was to compare SICI, cortical thickness, and cortical area in subjects with idiopathic mild to moderate RLS/WED, severe to very severe RLS/WED, and controls.

METHODS

The severity of RLS/WED was assessed by the International Restless Legs Syndrome Severity Scale (IRLSS). SICI and 3T magnetic resonance imaging (MRI) data of subjects with RLS/WED and controls were compared. A receiver operating characteristic curve for SICI was designed for discrimination of participants with RLS/WED from controls. Cortical thickness and area were assessed by automated surface-based analysis.

RESULTS

SICI was significantly reduced in patients with mild to moderate and severe to very severe RLS/WED, compared to controls (one-way analysis of variance: F = 9.62, p < 0.001). Receiver operating characteristic curve analysis predicted RLS/WED when SICI was above 35% (area under the curve = 0.79, 95% CI 0.67-0.91, p < 0.001). Analyses of the whole brain and of regions of interest did not reveal differences in gray matter thickness or area between controls and subjects with RLS/WED.

CONCLUSION

SICI is an accurate cortical biomarker that can support the diagnosis of RLS/WED even in subjects with mild symptoms, but cortical thickness and area were not useful for discriminating subjects with this condition from controls.

The duration of the cortical silent period is not abnormal in Restless Legs Syndrome/Willis-Ekbom Disease

OBJECTIVE

To compare the duration of the cortical silent period (CSP) measured in a hand muscle in subjects with primary Restless Legs Syndrome (RLS/WED) and controls, using four different methods of analysis.

METHODS

The CSP to transcranial magnetic stimulation of the dominant motor cortex was assessed in the abductor digiti minimi of 33 subjects with RLS/WED and 24 controls. CSP duration was measured by an automated and three visually-guided methods.

RESULTS

There were significant differences between absolute values of CSP duration according to the method of analysis. However, irrespectively of the method used for CSP assessment, no differences were found between measurements performed in subjects with RLS/WED and subjects from the control group.

CONCLUSIONS

Absolute values of CSP durations analyzed by different methods should not be directly compared, because significantly different results can be obtained from the same data set.

SIGNIFICANCE

The CSP assessed from a hand muscle is unlikely to be a biomarker of primary RLS/WED. Our results highlight the importance of standardizing the definition of CSP onset and offset, as well as of describing in detail the methodology chosen to record and measure CSP duration, in order to enable comparisons between studies.