Supporting equitable and responsible highway safety improvement funding allocation strategies - Why AI prediction biases matter

The existing methodologies for allocating highway safety improvement funding closely rely on the utilization of crash prediction models. Specifically, these models produce predictions that estimate future crash hazard levels in different geographical areas, which subsequently support the future funding allocation strategies. In recent years, there is a burgeoning interest in applying artificial intelligence (AI)-based models to perform crash prediction tasks. Despite the remarkable accuracy of these AI-based crash prediction models, they have been observed to yield biased prediction outcomes across areas of different socioeconomic statuses. These biases are primarily attributed to the inherent measurement and representation biases of AI-based prediction models. More precisely, measurement bias arises from the selection of target variables to reflect crash hazard levels, while representation bias results from the issue of imbalanced number of samples representing areas with different socioeconomic statuses within the dataset. Consequently, these biased prediction outcomes have the potential to perpetuate an unfair allocation of funding resources, contributing to worsen social inequality over time. Drawing upon a real-world case study in North Carolina, this study designs an AI-based crash prediction model that utilizes previous sociodemographic and crash-related variables to predict future severe crash rate of each area to reflect the crash hazardous level. By incorporating a fair regression framework, this study endeavors to transform the crash prediction model to become both fair and accurate, aiming to support equitable and responsible safety improvement funding allocation strategies.

Systematic Review of Health Economic Evaluations Focused on Artificial Intelligence in Healthcare: The Tortoise and the Cheetah

OBJECTIVES

This study aimed to systematically review recent health economic evaluations (HEEs) of artificial intelligence (AI) applications in healthcare. The aim was to discuss pertinent methods, reporting quality and challenges for future implementation of AI in healthcare, and additionally advise future HEEs.

METHODS

A systematic literature review was conducted in 2 databases (PubMed and Scopus) for articles published in the last 5 years. Two reviewers performed independent screening, full-text inclusion, data extraction, and appraisal. The Consolidated Health Economic Evaluation Reporting Standards and Philips checklist were used for the quality assessment of included studies.

RESULTS

A total of 884 unique studies were identified; 20 were included for full-text review, covering a wide range of medical specialties and care pathway phases. The most commonly evaluated type of AI was automated medical image analysis models (n = 9, 45%). The prevailing health economic analysis was cost minimization (n = 8, 40%) with the costs saved per case as preferred outcome measure. A total of 9 studies (45%) reported model-based HEEs, 4 of which applied a time horizon >1 year. The evidence supporting the chosen analytical methods, assessment of uncertainty, and model structures was underreported. The reporting quality of the articles was moderate as on average studies reported on 66% of Consolidated Health Economic Evaluation Reporting Standards items.

CONCLUSIONS

HEEs of AI in healthcare are limited and often focus on costs rather than health impact. Surprisingly, model-based long-term evaluations are just as uncommon as model-based short-term evaluations. Consequently, insight into the actual benefits offered by AI is lagging behind current technological developments.

Assessing the Economic Value of Clinical Artificial Intelligence: Challenges and Opportunities

OBJECTIVES

Clinical artificial intelligence (AI) is a novel technology, and few economic evaluations have focused on it to date. Before its wider implementation, it is important to highlight the aspects of AI that challenge traditional health technology assessment methods.

METHODS

We used an existing broad value framework to assess potential ways AI can provide good value for money. We also developed a rubric of how economic evaluations of AI should vary depending on the case of its use.

RESULTS

We found that the measurement of core elements of value-health outcomes and cost-are complicated by AI because its generalizability across different populations is often unclear and because its use may necessitate reconfigured clinical processes. Clinicians' productivity may improve when AI is used. If poorly implemented though, AI may also cause clinicians' workload to increase. Some AI has been found to exacerbate health disparities. Nevertheless, AI may promote equity by expanding access to medical care and, when properly trained, providing unbiased diagnoses and prognoses. The approach to assessment of AI should vary based on its use case: AI that creates new clinical possibilities can improve outcomes, but regulation and evidence collection may be difficult; AI that extends clinical expertise can reduce disparities and lower costs but may result in overuse; and AI that automates clinicians' work can improve productivity but may reduce skills.

CONCLUSIONS

The potential uses of clinical AI create challenges for health technology assessment methods originally developed for pharmaceuticals and medical devices. Health economists should be prepared to examine data collection and methods used to train AI, as these may impact its future value.

Translatability Analysis of National Institutes of Health-Funded Biomedical Research That Applies Artificial Intelligence

IMPORTANCE

Despite the rapid growth of interest and diversity in applications of artificial intelligence (AI) to biomedical research, there are limited objective ways to characterize the potential for use of AI in clinical practice.

OBJECTIVE

To examine what types of medical AI have the greatest estimated translational impact (ie, ability to lead to development that has measurable value for human health) potential.

DESIGN, SETTING, AND PARTICIPANTS

In this cohort study, research grants related to AI awarded between January 1, 1985, and December 31, 2020, were identified from a National Institutes of Health (NIH) award database. The text content for each award was entered into a Natural Language Processing (NLP) clustering algorithm. An NIH database was also used to extract citation data, including the number of citations and approximate potential to translate (APT) score for published articles associated with the granted awards to create proxies for translatability.

EXPOSURES

Unsupervised assignment of AI-related research awards to application topics using NLP.

MAIN OUTCOMES AND MEASURES

Annualized citations per $1 million funding (ACOF) and average APT score for award-associated articles, grouped by application topic. The APT score is a machine-learning based metric created by the NIH Office of Portfolio Analysis that quantifies the likelihood of future citation by a clinical article.

RESULTS

A total of 16 629 NIH awards related to AI were included in the analysis, and 75 applications of AI were identified. Total annual funding for AI grew from $17.4 million in 1985 to $1.43 billion in 2020. By average APT, interpersonal communication technologies (0.488; 95% CI, 0.472-0.504) and population genetics (0.463; 95% CI, 0.453-0.472) had the highest translatability; environmental health (ACOF, 1038) and applications focused on the electronic health record (ACOF, 489) also had high translatability. The category of applications related to biochemical analysis was found to have low translatability by both metrics (average APT, 0.393; 95% CI, 0.388-0.398; ACOF, 246).

CONCLUSIONS AND RELEVANCE

Based on this study's findings, data on grants from the NIH can apparently be used to identify and characterize medical applications of AI to understand changes in academic productivity, funding support, and potential for translational impact. This method may be extended to characterize other research domains.

Cardiovascular RNA markers and artificial intelligence may improve COVID-19 outcome: a position paper from the EU-CardioRNA COST Action CA17129

The coronavirus disease 2019 (COVID-19) pandemic has been as unprecedented as unexpected, affecting more than 105 million people worldwide as of 8 February 2020 and causing more than 2.3 million deaths according to the World Health Organization (WHO). Not only affecting the lungs but also provoking acute respiratory distress, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is able to infect multiple cell types including cardiac and vascular cells. Hence a significant proportion of infected patients develop cardiac events, such as arrhythmias and heart failure. Patients with cardiovascular comorbidities are at highest risk of cardiac death. To face the pandemic and limit its burden, health authorities have launched several fast-track calls for research projects aiming to develop rapid strategies to combat the disease, as well as longer-term projects to prepare for the future. Biomarkers have the possibility to aid in clinical decision-making and tailoring healthcare in order to improve patient quality of life. The biomarker potential of circulating RNAs has been recognized in several disease conditions, including cardiovascular disease. RNA biomarkers may be useful in the current COVID-19 situation. The discovery, validation, and marketing of novel biomarkers, including RNA biomarkers, require multi-centre studies by large and interdisciplinary collaborative networks, involving both the academia and the industry. Here, members of the EU-CardioRNA COST Action CA17129 summarize the current knowledge about the strain that COVID-19 places on the cardiovascular system and discuss how RNA biomarkers can aid to limit this burden. They present the benefits and challenges of the discovery of novel RNA biomarkers, the need for networking efforts, and the added value of artificial intelligence to achieve reliable advances.

Cost Effectiveness of an Electrocardiographic Deep Learning Algorithm to Detect Asymptomatic Left Ventricular Dysfunction

OBJECTIVE

To evaluate the cost-effectiveness of an artificial intelligence electrocardiogram (AI-ECG) algorithm under various clinical and cost scenarios when used for universal screening at age 65.

PATIENTS AND METHODS

We used decision analytic modeling to perform a cost-effectiveness analysis of the use of AI-ECG to screen for asymptomatic left ventricular dysfunction (ALVD) once at age 65 compared with no screening. This screening consisted of an initial screening decision tree and subsequent construction of a Markov model. One-way sensitivity analysis on various disease and cost parameters to evaluate cost-effectiveness at both $50,000 per quality-adjusted life year (QALY) and $100,000 per QALY willingness-to-pay threshold.

RESULTS

We found that for universal screening at age 65, the novel AI-ECG algorithm would cost $43,351 per QALY gained, test performance, disease characteristics, and testing cost parameters significantly affect cost-effectiveness, and screening at ages 55 and 75 would cost $48,649 and $52,072 per QALY gained, respectively. Overall, under most of the clinical scenarios modeled, coupled with its robust test performance in both testing and validation cohorts, screening with the novel AI-ECG algorithm appears to be cost-effective at a willingness-to-pay threshold of $50,000.

CONCLUSION

Universal screening for ALVD with the novel AI-ECG appears to be cost-effective under most clinical scenarios with a cost of <$50,000 per QALY. Cost-effectiveness is particularly sensitive to both the probability of disease progression and the cost of screening and downstream testing. To improve cost-effectiveness modeling, further study of the natural progression and treatment of ALVD and external validation of AI-ECG should be undertaken.

Artificial intelligence-the future is now

The pros and cons of artificial intelligence in assisted reproductive technology are presented.

Firm information disclosure environment and R&D investment: Evidence from Internet penetration

Guided by the conviction that "Clear waters and green mountains are as good as mountains of gold and silver", China highly values sustainable economic and social development through innovation and Internet technology. Regression analysis is performed to examine the impact of corporate information disclosure environment proxied by the Internet penetration rate on innovation. Leveraging from the city-level Internet penetration rates data in China from 2003 to 2017, this study gets the following findings: (1) Firms headquartered in cities with high Internet penetration rates tend to be more innovative, i.e. they invest more in research and development. (2) This result is supported by several robustness checks, such as alternative measures of key variables, alternative empirical specifications, and tests to mitigate identification concerns. (3) "financing constraint" and "tolerance of innovation failure" are two channels that influence firms' innovative endeavors. (4) Additional tests show that Internet penetration rates facilitate a firm's output efficiency of innovation input, total factor productivity, and human capital environment for innovation. The above conclusions not only enrich the relevant literature on the influencing factors of corporate innovation from the perspective of the firm information disclosure environment but also provide an important reference for further understanding the positive role of macro technology development on social and economic development.

Covid-19 driven advances in automation and artificial intelligence risk exacerbating economic inequality

Anton Korinek and Joseph E Stiglitz make the case for a deliberate effort to steer technological advances in a direction that enhances the role of human workers

Four ethical priorities for neurotechnologies and AI

Artificial intelligence and brain-computer interfaces must respect and preserve people’s privacy, identity, agency and equality, say Rafael Yuste, Sara Goering and colleagues.

A heuristic placement selection of live virtual machine migration for energy-saving in cloud computing environment

The field of live VM (virtual machine) migration has been a hotspot problem in green cloud computing. Live VM migration problem is divided into two research aspects: live VM migration mechanism and live VM migration policy. In the meanwhile, with the development of energy-aware computing, we have focused on the VM placement selection of live migration, namely live VM migration policy for energy saving. In this paper, a novel heuristic approach PS-ES is presented. Its main idea includes two parts. One is that it combines the PSO (particle swarm optimization) idea with the SA (simulated annealing) idea to achieve an improved PSO-based approach with the better global search's ability. The other one is that it uses the Probability Theory and Mathematical Statistics and once again utilizes the SA idea to deal with the data obtained from the improved PSO-based process to get the final solution. And thus the whole approach achieves a long-term optimization for energy saving as it has considered not only the optimization of the current problem scenario but also that of the future problem. The experimental results demonstrate that PS-ES evidently reduces the total incremental energy consumption and better protects the performance of VM running and migrating compared with randomly migrating and optimally migrating. As a result, the proposed PS-ES approach has capabilities to make the result of live VM migration events more high-effective and valuable.

A low computation cost method for seizure prediction

The dynamic changes of electroencephalograph (EEG) signals in the period prior to epileptic seizures play a major role in the seizure prediction. This paper proposes a low computation seizure prediction algorithm that combines a fractal dimension with a machine learning algorithm. The presented seizure prediction algorithm extracts the Higuchi fractal dimension (HFD) of EEG signals as features to classify the patient's preictal or interictal state with Bayesian linear discriminant analysis (BLDA) as a classifier. The outputs of BLDA are smoothed by a Kalman filter for reducing possible sporadic and isolated false alarms and then the final prediction results are produced using a thresholding procedure. The algorithm was evaluated on the intracranial EEG recordings of 21 patients in the Freiburg EEG database. For seizure occurrence period of 30 min and 50 min, our algorithm obtained an average sensitivity of 86.95% and 89.33%, an average false prediction rate of 0.20/h, and an average prediction time of 24.47 min and 39.39 min, respectively. The results confirm that the changes of HFD can serve as a precursor of ictal activities and be used for distinguishing between interictal and preictal epochs. Both HFD and BLDA classifier have a low computational complexity. All of these make the proposed algorithm suitable for real-time seizure prediction.

Leveraging business intelligence to make better decisions: Part I

Data is the new currency. Business intelligence tools will provide better performing practices with a competitive intelligence advantage that will separate the high performers from the rest of the pack. Given the investments of time and money into our data systems, practice leaders must work to take every advantage and look at the datasets as a potential goldmine of business intelligence decision tools. A fresh look at decision tools created from practice data will create efficiencies and improve effectiveness for end-users and managers.