What Complexity Science Predicts About the Potential of Artificial Intelligence/Machine Learning to Improve Primary Care.

Primary care physicians are likely both excited and apprehensive at the prospects for artificial intelligence (AI) and machine learning (ML). Complexity science may provide insight into which AI/ML applications will most likely affect primary care in the future. AI/ML has successfully diagnosed some diseases from digital images, helped with administrative tasks such as writing notes in the electronic record by converting voice to text, and organized information from multiple sources within a health care system. AI/ML has less successfully recommended treatments for patients with complicated single diseases such as cancer; or improved diagnosing, patient shared decision making, and treating patients with multiple comorbidities and social determinant challenges. AI/ML has magnified disparities in health equity, and almost nothing is known of the effect of AI/ML on primary care physician-patient relationships. An intervention in Victoria, Australia showed promise where an AI/ML tool was used only as an adjunct to complex medical decision making. Putting these findings in a complex adaptive system framework, AI/ML tools will likely work when its tasks are limited in scope, have clean data that are mostly linear and deterministic, and fit well into existing workflows. AI/ML has rarely improved comprehensive care, especially in primary care settings, where data have a significant number of errors and inconsistencies. Primary care should be intimately involved in AI/ML development, and its tools carefully tested before implementation; and unlike electronic health records, not just assumed that AI/ML tools will improve primary care work life, quality, safety, and person-centered clinical decision making.

Artificial intelligence research strategy of the United States: critical assessment and policy recommendations.

The foundations of Artificial Intelligence (AI), a field whose applications are of great use and concern for society, can be traced back to the early years of the second half of the 20th century. Since then, the field has seen increased research output and funding cycles followed by setbacks. The new millennium has seen unprecedented interest in AI progress and expectations with significant financial investments from the public and private sectors. However, the continual acceleration of AI capabilities and real-world applications is not guaranteed. Mainly, accountability of AI systems in the context of the interplay between AI and the broader society is essential for adopting AI systems via the trust placed in them. Continual progress in AI research and development (R&D) can help tackle humanity's most significant challenges to improve social good. The authors of this paper suggest that the careful design of forward-looking research policies serves a crucial function in avoiding potential future setbacks in AI research, development, and use. The United States (US) has kept its leading role in R&D, mainly shaping the global trends in the field. Accordingly, this paper presents a critical assessment of the US National AI R&D Strategic Plan and prescribes six recommendations to improve future research strategies in the US and around the globe.

Factors Contributing to SARS-CoV-2 Vaccine Hesitancy of Hispanic Population in Rio Grande Valley.

Hispanic communities have been disproportionately affected by economic disparities. These inequalities have put Hispanics at an increased risk for preventable health conditions. In addition, the CDC reports Hispanics to have 1.5× COVID-19 infection rates and low vaccination rates. This study aims to identify the driving factors for COVID-19 vaccine hesitancy of Hispanic survey participants in the Rio Grande Valley. Our analysis used machine learning methods to identify significant associations between medical, economic, and social factors impacting the uptake and willingness to receive the COVID-19 vaccine. A combination of three classification methods (i.e., logistic regression, decision trees, and support vector machines) was used to classify observations based on the value of the targeted responses received and extract a robust subset of factors. Our analysis revealed different medical, economic, and social associations that correlate to other target population groups (i.e., males and females). According to the analysis performed on males, the Matthews correlation coefficient (MCC) value was 0.972. An MCC score of 0.805 was achieved by analyzing females, while the analysis of males and females achieved 0.797. Specifically, several medical, economic factors, and sociodemographic characteristics are more prevalent in vaccine-hesitant groups, such as asthma, hypertension, mental health problems, financial strain due to COVID-19, gender, lack of health insurance plans, and limited test availability.

Advancing primary care with Artificial Intelligence and Machine Learning.

Primary care is the largest healthcare delivery platform in the US. Facing the Artificial Intelligence and Machine Learning technology (AI/ML) revolution, the primary care community would benefit from a roadmap revealing priority areas and opportunities for developing and integrating AI/ML-driven clinical tools. This article presents a framework that identifies five domains for AI/ML integration in primary care to support care delivery transformation and achieve the Quintuple Aims of the healthcare system. We concluded that primary care plays a critical role in developing, introducing, implementing, and monitoring AI/ML tools in healthcare and must not be overlooked as AI/ML transforms healthcare.

EDGE20: A Cross Spectral Evaluation Dataset for Multiple Surveillance Problems.

Surveillance-related datasets that have been released in recent years focus only on one specific problem at a time (e.g., pedestrian detection, face detection, or face recognition), while most of them were collected using visible spectrum (VIS) cameras. Even though some cross-spectral datasets were presented in the past, they were acquired in a constrained setup, which limited the performance of methods for the aforementioned problems under a cross-spectral setting. This work introduces a new dataset, named EDGE19, that can be used in addressing the problems of pedestrian detection, face detection, and face recognition in images captured using trail cameras under the VIS and NIR spectra. Data acquisition was performed in an outdoor environment, during both day and night, under unconstrained acquisition conditions. The collection of images is accompanied by a rich set of annotations, consisting of person and facial bounding boxes, unique subject identifiers, and labels that characterize facial images as frontal, profile, or back faces. Moreover, the performance of several state-of-the-art methods was evaluated for each of the scenarios covered by our dataset. The baseline results we obtained highlight the difficulty of current methods in the tasks of cross-spectral pedestrian detection, face detection, and face recognition due to unconstrained conditions, including low resolution, pose variation, illumination variation, occlusions, and motion blur.

An exploratory decision tree analysis to predict physical activity compliance rates in breast cancer survivors.

Background: The study of physical activity in cancer survivors has been limited to one cause, one effect relationships. In this exploratory study, we used recursive partitioning to examine multiple correlates that influence physical activity compliance rates in cancer survivors. Methods: African American breast cancer survivors (N = 267, Mean age = 54 years) participated in an online survey that examined correlates of physical activity. Recursive partitioning (RP) was used to examine complex and nonlinear associations between sociodemographic, medical, cancer-related, theoretical, and quality of life indicators. Results: Recursive partitioning revealed five distinct groups. Compliance with physical activity guidelines was highest (82% met guidelines) among survivors who reported higher mean action planning scores (P < 0.001) and lower mean barriers to physical activity (P = 0.035). Compliance with physical activity guidelines was lowest (9% met guidelines) among survivors who reported lower mean action and coping (P = 0.002) planning scores. Similarly, lower mean action planning scores and poor advanced lower functioning (P = 0.034), even in the context of higher coping planning scores, resulted in low physical activity compliance rates (13% met guidelines). Subsequent analyses revealed that body mass index (P = 0.019) and number of comorbidities (P = 0.003) were lowest in those with the highest compliance rates. Conclusion: Our findings support the notion that multiple factors determine physical activity compliance rates in African American breast cancer survivors. Interventions that encourage action and coping planning and reduce barriers in the context of addressing function limitations may increase physical activity compliance rates.

Machine Learning Outperforms ACC / AHA CVD Risk Calculator in MESA.

Background Studies have demonstrated that the current US guidelines based on American College of Cardiology/American Heart Association (ACC/AHA) Pooled Cohort Equations Risk Calculator may underestimate risk of atherosclerotic cardiovascular disease ( CVD ) in certain high-risk individuals, therefore missing opportunities for intensive therapy and preventing CVD events. Similarly, the guidelines may overestimate risk in low risk populations resulting in unnecessary statin therapy. We used Machine Learning ( ML ) to tackle this problem. Methods and Results We developed a ML Risk Calculator based on Support Vector Machines ( SVM s) using a 13-year follow up data set from MESA (the Multi-Ethnic Study of Atherosclerosis) of 6459 participants who were atherosclerotic CVD-free at baseline. We provided identical input to both risk calculators and compared their performance. We then used the FLEMENGHO study (the Flemish Study of Environment, Genes and Health Outcomes) to validate the model in an external cohort. ACC / AHA Risk Calculator, based on 7.5% 10-year risk threshold, recommended statin to 46.0%. Despite this high proportion, 23.8% of the 480 "Hard CVD " events occurred in those not recommended statin, resulting in sensitivity 0.76, specificity 0.56, and AUC 0.71. In contrast, ML Risk Calculator recommended only 11.4% to take statin, and only 14.4% of "Hard CVD " events occurred in those not recommended statin, resulting in sensitivity 0.86, specificity 0.95, and AUC 0.92. Similar results were found for prediction of "All CVD " events. Conclusions The ML Risk Calculator outperformed the ACC/AHA Risk Calculator by recommending less drug therapy, yet missing fewer events. Additional studies are underway to validate the ML model in other cohorts and to explore its ability in short-term CVD risk prediction.

An Overview and Empirical Comparison of Distance Metric Learning Methods.

In this paper, we first offer an overview of advances in the field of distance metric learning. Then, we empirically compare selected methods using a common experimental protocol. The number of distance metric learning algorithms proposed keeps growing due to their effectiveness and wide application. However, existing surveys are either outdated or they focus only on a few methods. As a result, there is an increasing need to summarize the obtained knowledge in a concise, yet informative manner. Moreover, existing surveys do not conduct comprehensive experimental comparisons. On the other hand, individual distance metric learning papers compare the performance of the proposed approach with only a few related methods and under different settings. This highlights the need for an experimental evaluation using a common and challenging protocol. To this end, we conduct face verification experiments, as this task poses significant challenges due to varying conditions during data acquisition. In addition, face verification is a natural application for distance metric learning because the encountered challenge is to define a distance function that: 1) accurately expresses the notion of similarity for verification; 2) is robust to noisy data; 3) generalizes well to unseen subjects; and 4) scales well with the dimensionality and number of training samples. In particular, we utilize well-tested features to assess the performance of selected methods following the experimental protocol of the state-of-the-art database labeled faces in the wild. A summary of the results is presented along with a discussion of the insights obtained and lessons learned by employing the corresponding algorithms.

A physics-based intravascular ultrasound image reconstruction method for lumen segmentation.

Intravascular ultrasound (IVUS) refers to the medical imaging technique consisting of a miniaturized ultrasound transducer located at the tip of a catheter that can be introduced in the blood vessels providing high-resolution, cross-sectional images of their interior. Current methods for the generation of an IVUS image reconstruction from radio frequency (RF) data do not account for the physics involved in the interaction between the IVUS ultrasound signal and the tissues of the vessel. In this paper, we present a novel method to generate an IVUS image reconstruction based on the use of a scattering model that considers the tissues of the vessel as a distribution of three-dimensional point scatterers. We evaluated the impact of employing the proposed IVUS image reconstruction method in the segmentation of the lumen/wall interface on 40MHz IVUS data using an existing automatic lumen segmentation method. We compared the results with those obtained using the B-mode reconstruction on 600 randomly selected frames from twelve pullback sequences acquired from rabbit aortas and different arteries of swine. Our results indicate the feasibility of employing the proposed IVUS image reconstruction for the segmentation of the lumen.

A novel method for imaging the pharmacological effects of antibiotic treatment on Clostridium difficile.

Clostridium difficile is a significant cause of nosocomial-acquired infection that results in severe diarrhea and can lead to mortality. Treatment options for C. difficile infection (CDI) are limited, however, new antibiotics are being developed. Current methods for determining efficacy of experimental antibiotics on C. difficile involve antibiotic killing rates and do not give insight into the drug's pharmacologic effects. Considering this, we hypothesized that by using scanning electron microscopy (SEM) in tandem to drug killing curves, we would be able to determine efficacy and visualize the phenotypic response to drug treatment. To test this hypothesis, supraMIC kill curves were conducted using vancomycin, metronidazole, fidaxomicin, and ridinilazole. Following collection, cells were either plated or imaged using a scanning electron microscope (SEM). Consistent with previous reports, we found that the tested antibiotics had significant bactericidal activity at supraMIC concentrations. By SEM imaging and using a semi-automatic pipeline for image analysis, we were able to determine that vancomycin and to a lesser extent fidaxomicin and ridinilazole significantly affected the cell wall, whereas metronidazole, fidaxomicin, and ridinilazole had significant effects on cell length suggesting a metabolic effect. While the phenotypic response to drug treatment has not been documented previously in this manner, the results observed are consistent with the drug's mechanism of action. These techniques demonstrate the versatility and reliability of imaging and measurements that could be applied to other experimental compounds. We believe the strategies laid out here are vital for characterizing new antibiotics in development for treating CDI.

Multi-scale segmentation of neurons based on one-class classification.

BACKGROUND: High resolution multiphoton and confocal microscopy has allowed the acquisition of large amounts of data to be analyzed by neuroscientists. However, manual processing of these images has become infeasible. Thus, there is a need to create automatic methods for the morphological reconstruction of 3D neuronal image stacks. NEW METHOD: An algorithm to extract the 3D morphology from a neuron is presented. The main contribution of the paper is the segmentation of the neuron from the background. Our segmentation method is based on one-class classification where the 3D image stack is analyzed at different scales. First, a multi-scale approach is proposed to compute the Laplacian of the 3D image stack. The Laplacian is used to select a training set consisting of background points. A decision function is learned for each scale from the training set that allows determining how similar an unlabeled point is to the points in the background class. Foreground points (dendrites and axons) are assigned as those points that are rejected as background. Finally, the morphological reconstruction of the neuron is extracted by applying a state-of-the-art centerline tracing algorithm on the segmentation. RESULTS: Quantitative and qualitative results on several datasets demonstrate the ability of our algorithm to accurately and robustly segment and trace neurons. COMPARISON WITH EXISTING METHOD(S): Our method was compared to state-of-the-art neuron tracing algorithms. CONCLUSIONS: Our approach allows segmentation of thin and low contrast dendrites that are usually difficult to segment. Compared to our previous approach, this algorithm is more accurate and much faster.

Automatic 2.5-D Facial Landmarking and Emotion Annotation for Social Interaction Assistance.

People with low vision, Alzheimer's disease, and autism spectrum disorder experience difficulties in perceiving or interpreting facial expression of emotion in their social lives. Though automatic facial expression recognition (FER) methods on 2-D videos have been extensively investigated, their performance was constrained by challenges in head pose and lighting conditions. The shape information in 3-D facial data can reduce or even overcome these challenges. However, high expenses of 3-D cameras prevent their widespread use. Fortunately, 2.5-D facial data from emerging portable RGB-D cameras provide a good balance for this dilemma. In this paper, we propose an automatic emotion annotation solution on 2.5-D facial data collected from RGB-D cameras. The solution consists of a facial landmarking method and a FER method. Specifically, we propose building a deformable partial face model and fit the model to a 2.5-D face for localizing facial landmarks automatically. In FER, a novel action unit (AU) space-based FER method has been proposed. Facial features are extracted using landmarks and further represented as coordinates in the AU space, which are classified into facial expressions. Evaluated on three publicly accessible facial databases, namely EURECOM, FRGC, and Bosphorus databases, the proposed facial landmarking and expression recognition methods have achieved satisfactory results. Possible real-world applications using our algorithms have also been discussed.

Automatic Morphological Reconstruction of Neurons from Multiphoton and Confocal Microscopy Images Using 3D Tubular Models.

The challenges faced in analyzing optical imaging data from neurons include a low signal-to-noise ratio of the acquired images and the multiscale nature of the tubular structures that range in size from hundreds of microns to hundreds of nanometers. In this paper, we address these challenges and present a computational framework for an automatic, three-dimensional (3D) morphological reconstruction of live nerve cells. The key aspects of this approach are: (i) detection of neuronal dendrites through learning 3D tubular models, and (ii) skeletonization by a new algorithm using a morphology-guided deformable model for extracting the dendritic centerline. To represent the neuron morphology, we introduce a novel representation, the Minimum Shape-Cost (MSC) Tree that approximates the dendrite centerline with sub-voxel accuracy and demonstrate the uniqueness of such a shape representation as well as its computational efficiency. We present extensive quantitative and qualitative results that demonstrate the accuracy and robustness of our method.

Improved automatic centerline tracing for dendritic and axonal structures.

Centerline tracing in dendritic structures acquired from confocal images of neurons is an essential tool for the construction of geometrical representations of a neuronal network from its coarse scale up to its fine scale structures. In this paper, we propose an algorithm for centerline extraction that is both highly accurate and computationally efficient. The main novelties of the proposed method are (1) the use of a small set of Multiscale Isotropic Laplacian filters, acting as self-steerable filters, for a quick and efficient binary segmentation of dendritic arbors and axons; (2) an automated centerline seed points detection method based on the application of a simple 3D finite-length filter. The performance of this algorithm, which is validated on data from the DIADEM set appears to be very competitive when compared with other state-of-the-art algorithms.

Can We Do Better in Unimodal Biometric Systems? A Rank-Based Score Normalization Framework.

Biometric systems use score normalization techniques and fusion rules to improve recognition performance. The large amount of research on score fusion for multimodal systems raises an important question: can we utilize the available information from unimodal systems more effectively? In this paper, we present a rank-based score normalization framework that addresses this problem. Specifically, our approach consists of three algorithms: 1) partition the matching scores into subsets and normalize each subset independently; 2) utilize the gallery versus gallery matching scores matrix (i.e., gallery-based information); and 3) dynamically augment the gallery in an online fashion. We invoke the theory of stochastic dominance along with results of prior research to demonstrate when and why our approach yields increased performance. Our framework: 1) can be used in conjunction with any score normalization technique and any fusion rule; 2) is amenable to parallel programming; and 3) is suitable for both verification and open-set identification. To assess the performance of our framework, we use the UHDB11 and FRGC v2 face datasets. Specifically, the statistical hypothesis tests performed illustrate that the performance of our framework improves as we increase the number of samples per subject. Furthermore, the corresponding statistical analysis demonstrates that increased separation between match and nonmatch scores is obtained for each probe. Besides the benefits and limitations highlighted by our experimental evaluation, results under optimal and pessimal conditions are also presented to offer better insights.

PDM-ENLOR for segmentation of mouse brain gene expression images.

Statistical shape models, such as Active Shape Models (ASMs), suffer from their inability to represent a large range of variations of a complex shape and to account for the large errors in detection of (point) landmarks. We propose a method, PDM-ENLOR (Point Distribution Model-based ENsemble of LOcal Regressors), that overcomes these limitations by locating each landmark individually using an ensemble of local regression models and appearance cues from selected landmarks. We first detect a set of reference landmarks which were selected based on their saliency during training. For each landmark, an ensemble of regressors is built. From the locations of the detected reference landmarks, each regressor infers a candidate location for that landmark using local geometric constraints, encoded by a point distribution model (PDM). The final location of that point is determined as a weighted linear combination, whose coefficients are learned from the training data, of candidates proposed by its ensemble's component regressors. We use multiple subsets of reference landmarks as explanatory variables for the component regressors to provide varying degrees of locality for the models in each ensemble. This helps our ensemble model to capture a larger range of shape variations as compared to a single PDM. We demonstrate the advantages of our method on the challenging problem of segmenting gene expression images of mouse brain. The overall mean and standard deviation of the Dice coefficient overlap over all 14 anatomical regions and all 100 test images were (88.1 ± 9.5)%.

Identification of environmental chemicals that induce yolk malabsorption in zebrafish using automated image segmentation.

Environmental factors affecting nutrient availability during development can cause predisposition to diseases later in life. To identify chemicals in the environment capable of altering nutrient mobilization, we analyzed yolk malabsorption in the zebrafish embryo, which relies on maternally-derived yolk for nutrition during its first week of life. Embryos of the transgenic zebrafish line HGn50D, which fluoresce in the yolk syncytial layer, were exposed from two to five days post fertilization to different chemicals. We developed a software package to automatically and accurately segment and quantify the area of the fluorescing yolk in images captured at the end of the treatment period. Based on this quantification, we found that prochloraz decreased yolk absorption, while butralin, tetrabromobisphenol A, tetrachlorobisphenol A and tributyltin increased yolk absorption. Given the number and variety of industrial chemicals in commerce today, development of automated image processing to perform high-speed quantitative analysis of biological effects is an important step for enabling high throughput screening to identify chemicals altering nutrient absorption.