Month: April 2024

Advocating for population health: The role of public health practitioners in the age of artificial intelligence

Latest Results

Abstract

Over the past decade, artificial intelligence (AI) has begun to transform Canadian organizations, driven by the promise of improved efficiency, better decision-making, and enhanced client experience. While AI holds great opportunities, there are also near-term impacts on the determinants of health and population health equity that are already emerging. If adoption is unregulated, there is a substantial risk that health inequities could be exacerbated through intended or unintended biases embedded in AI systems. New economic opportunities could be disproportionately leveraged by already privileged workers and owners of AI systems, reinforcing prevailing power dynamics. AI could also detrimentally affect population well-being by replacing human interactions rather than fostering social connectedness. Furthermore, AI-powered health misinformation could undermine effective public health communication. To respond to these challenges, public health must assess and report on the health equity impacts of AI, inform implementation to reduce health inequities, and facilitate intersectoral partnerships to foster development of policies and regulatory frameworks to mitigate risks. This commentary highlights AI’s near-term risks for population health to inform a public health response.

Developing a hierarchical model for unraveling conspiracy theories

Latest Results

Abstract

A conspiracy theory (CT) suggests covert groups or powerful individuals secretly manipulate events. Not knowing about existing conspiracy theories could make one more likely to believe them, so this work aims to compile a list of CTs shaped as a tree that is as comprehensive as possible. We began with a manually curated ‘tree’ of CTs from academic papers and Wikipedia. Next, we examined 1769 CT-related articles from four fact-checking websites, focusing on their core content, and used a technique called Keyphrase Extraction to label the documents. This process yielded 769 identified conspiracies, each assigned a label and a family name. The second goal of this project was to detect whether an article is a conspiracy theory, so we built a binary classifier with our labeled dataset. This model uses a transformer-based machine learning technique and is pre-trained on a large corpus called RoBERTa, resulting in an F1 score of 87%. This model helps to identify potential conspiracy theories in new articles. We used a combination of clustering (HDBSCAN) and a dimension reduction technique (UMAP) to assign a label from the tree to these new articles detected as conspiracy theories. We then labeled these groups accordingly to help us match them to the tree. These can lead us to detect new conspiracy theories and expand the tree using computational methods. We successfully generated a tree of conspiracy theories and built a pipeline to detect and categorize conspiracy theories within any text corpora. This pipeline gives us valuable insights through any databases formatted as text.

Developing a hierarchical model for unraveling conspiracy theories

Latest Results

Abstract

A conspiracy theory (CT) suggests covert groups or powerful individuals secretly manipulate events. Not knowing about existing conspiracy theories could make one more likely to believe them, so this work aims to compile a list of CTs shaped as a tree that is as comprehensive as possible. We began with a manually curated ‘tree’ of CTs from academic papers and Wikipedia. Next, we examined 1769 CT-related articles from four fact-checking websites, focusing on their core content, and used a technique called Keyphrase Extraction to label the documents. This process yielded 769 identified conspiracies, each assigned a label and a family name. The second goal of this project was to detect whether an article is a conspiracy theory, so we built a binary classifier with our labeled dataset. This model uses a transformer-based machine learning technique and is pre-trained on a large corpus called RoBERTa, resulting in an F1 score of 87%. This model helps to identify potential conspiracy theories in new articles. We used a combination of clustering (HDBSCAN) and a dimension reduction technique (UMAP) to assign a label from the tree to these new articles detected as conspiracy theories. We then labeled these groups accordingly to help us match them to the tree. These can lead us to detect new conspiracy theories and expand the tree using computational methods. We successfully generated a tree of conspiracy theories and built a pipeline to detect and categorize conspiracy theories within any text corpora. This pipeline gives us valuable insights through any databases formatted as text.

Controlling factors of latitudinal distribution of dissolved organic matter in the upper layers of the Indian Ocean

Latest Results

Abstract

We studied chromophoric (CDOM) and fluorescent (FDOM) dissolved organic matter (DOM) and dissolved organic carbon in surface waters to determine the factors controlling the geographical distribution of DOM along two meridional transects in the Indian Ocean. For CDOM, we calculated the absorption coefficients, spectral slope, and absorption coefficient ratio from the observed absorption spectra. For FDOM, we calculated the biological (BIX) and humification (HIX) indices from the excitation emission matrices (EEMs); parallel factor analysis of the EEMs identified three fluorescent components, i.e., two humic-like and one protein-like. Using these DOM parameters, a factor analysis extracted fewer latent variables than the observed variables to account for the geographical distributions. We obtained three factors (F1, F2, and F3), which explained ~ 84% of the variance in the observed data. From the factor loadings, F1, F2, and F3 were interpreted as the effects of net primary production-derived DOM and its horizontal transport, photodegradation, and vertical transport by physical processes. We characterized seven marine biogeochemical provinces by factor scores. F1 scores gradually decreased from the northernmost to the Antarctic province, with a small maximum around the subtropical front. F2 scores were highest in the subtropical province and decreased in both the northward and southward directions. F3 scores were high in the Antarctic and northernmost provinces, and lowest in the subtropical province. Only BIX was insufficiently explained by these factors. BIX was highest in the northern part of the subtropical province, where photodegradation of DOM was the most intense. This suggests that the possible interaction between photodegradation, autochthonous production, and reworking by heterotrophic bacteria of DOM occurs in the subtropical province.

A comprehensive review on deep cardiovascular disease detection approaches: its datasets, image modalities and methods

Latest Results

Abstract

Early and accurate cardiovascular disease detection is a very crucial task to lower the mortality rate of a patient with a diagnosis of cardiovascular disease. Deep learning approaches are working effectively in discriminating or extracting important features from cardiovascular images to detect the disease. The objective of this paper is to deliver a comprehensive and detailed review article based on comparative analysis of imaging modalities, datasets, and deep learning architectures. The paper focuses to do an in-depth review on advances of deep architectural networks with relevant feature based characteristics used in cardiovascular disease (CVD) diagnosis. The novelty of the paper lies in including characteristics and challenges of cardiovascular imaging, their milestones in deep learning techniques, taxonomy of cardiovascular imaging with their purpose, dataset observation summarization, diagnosis strategies through deep learning and finally, deep learning architecture analysis. The performance of DL networks has been analyzed through CVD classification, segmentation and detection. It has been reported that AlexNet generates highest classification accuracy with 99%; for segmentation purpose, U-Net is the best technique with dice score 0.98 and in CVD detection, DenseNet 121, TR-Net, ResNet50 provide approximately 92% detection rate. At last, important findings are reported and identified as promising research directions for the future.

Few-shot classification with prototypical neural network for hospital flow recognition under uncertainty

Latest Results

Abstract

Accurately identifying and analyzing patient and personnel flow patterns within healthcare facilities is crucial for optimizing operational efficiency and delivering high-quality healthcare services. In this study, we propose a Prototypical Neural Network (PNN) tailored for few-shot learning, which effectively learns a representation space from limited labeled data. This enables efficient recognition of distinct characteristics within hospital flow footprints, ensuring examples from the same class are proximate while those from different classes are distant. Additionally, we introduce a synthetic sampling technique (SST) to address uncertainties and variations inherent in hospital personnel flow, thereby enhancing the robustness and performance of our flow recognition system. Through extensive simulation studies, we evaluate our approach and compare it against various classification methods, including support vector machine (SVM), random forest, naive Bayes classifier, residual neural network (ResNet), and fully connected neural network. The results showcase the superior performance of the proposed method, achieving an impressive accuracy of 99.17% in hospital flow footprint recognition. This outperforms classical methods, which range from 40.27% for fully connected neural networks to 80.55% for CNN. These findings underscore the efficacy of our method in recognizing hospital flow footprints, particularly in contexts characterized by uncertainty and variability.

Defining the structure–function relationship of specific lesions in early and advanced age-related macular degeneration

Latest Results

Abstract

The objective of this study is to define structure–function relationships of pathological lesions related to age-related macular degeneration (AMD) using microperimetry and multimodal retinal imaging. We conducted a cross-sectional study of 87 patients with AMD (30 eyes with early and intermediate AMD and 110 eyes with advanced AMD), compared to 33 normal controls (66 eyes) recruited from a single tertiary center. All participants had enface and cross-sectional optical coherence tomography (Heidelberg HRA-2), OCT angiography, color and infra-red (IR) fundus and microperimetry (MP) (Nidek MP-3) performed. Multimodal images were graded for specific AMD pathological lesions. A custom marking tool was used to demarcate lesion boundaries on corresponding enface IR images, and subsequently superimposed onto MP color fundus photographs with retinal sensitivity points (RSP). The resulting overlay was used to correlate pathological structural changes to zonal functional changes. Mean age of patients with early/intermediate AMD, advanced AMD and controls were 73(SD = 8.2), 70.8(SD = 8), and 65.4(SD = 7.7) years respectively. Mean retinal sensitivity (MRS) of both early/intermediate (23.1 dB; SD = 5.5) and advanced AMD (18.1 dB; SD = 7.8) eyes were significantly worse than controls (27.8 dB, SD = 4.3) (p < 0.01). Advanced AMD eyes had significantly more unstable fixation (70%; SD = 63.6), larger mean fixation area (3.9 mm2; SD = 3.0), and focal fixation point further away from the fovea (0.7 mm; SD = 0.8), than controls (29%; SD = 43.9; 2.6 mm2; SD = 1.9; 0.4 mm; SD = 0.3) (p ≤ 0.01). Notably, 22 fellow eyes of AMD eyes (25.7 dB; SD = 3.0), with no AMD lesions, still had lower MRS than controls (p = 0.04). For specific AMD-related lesions, end-stage changes such as fibrosis (5.5 dB, SD = 5.4 dB) and atrophy (6.2 dB, SD = 7.0 dB) had the lowest MRS; while drusen and pigment epithelial detachment (17.7 dB, SD = 8.0 dB) had the highest MRS. Peri-lesional areas (20.2 dB, SD = 7.6 dB) and surrounding structurally normal areas (22.2 dB, SD = 6.9 dB) of the retina with no AMD lesions still had lower MRS compared to controls (27.8 dB, SD = 4.3 dB) (p < 0.01). Our detailed topographic structure–function correlation identified specific AMD pathological changes associated with a poorer visual function. This can provide an added value to the assessment of visual function to optimize treatment outcomes to existing and potentially future novel therapies.

Environmental regulation, outward foreign direct investment, and China’s green total factor productivity

Latest Results

Abstract

Increasing green total factor productivity (GTFP) is currently the primary goal of sustainable development worldwide. GTFP not only reflects the efficiency of economic expansion but also encompasses resource consumption and pollution. This research enhances the current understanding of GTFP by indicating that aside from reverse technology spillovers, labor mobility, and changes in industrial structure, additional factors, such as environmental regulations, exert a dynamic function in shaping the influence of outward foreign direct investment (OFDI) on the GTFP of the home nation. The empirical findings indicate that OFDI has a single threshold effect on GTFP, and the negative effect increases with the reinforcing of environmental control. The main impact comes from home country’s changes in green technology (GTC) rather than changes in green efficiency. Additionally, environmental regulation has a positive moderating effect on OFDI, the moderating effect of environmental regulation in western regions is more pronounced in promoting the home country’s GTC. It is imperative to take into account regional variations and devise distinct policies for eastern, central, and western regions.

Demographic, social, and clinical aspects associated with access to COVID-19 health care in Pará province, Brazilian Amazon

Latest Results

Abstract

Internal social disparities in the Brazilian Amazon became more evident during the COVID-19 pandemic. The aim of this work was to examine the demographic, social and clinical factors associated with access to COVID-19 health care in Pará Province in the Brazilian Amazon. This was an observational, cross-sectional, analytical study using a quantitative method through an online survey conducted from May to August 2023. People were eligible to participate if they were current residents of Pará, 18-years-old or older, with self-reported diagnoses of COVID-19 through rapid or laboratory tests. Participants completed an electronic survey was developed using Research Electronic Data Capture (REDCap) software—The adapted questionnaire “COVID-19 Global Clinical Platform: Case Report Form for Post-COVID Condition”. Questions focused on access to COVID-19 treatment, demographic characteristics, COVID-19 vaccine and clinical characteristics. Respondent-driven sampling was applied to recruit participants. Multiple logistic regression was utilized to identify the associated factors. Overall, a total of 638 participants were included. The average age was 31.1 years. Access to COVID-19 health care was 68.65% (438/638). The participants most likely to access health care were those with moderate or severe COVID-19 (p = 0.000; OR: 19.8) and females (p = 0.001; OR: 1.99). Moreover, participants who used homemade tea or herbal medicines were less likely to receive health care for COVID-19 in health services (p = 0.002; OR: 0.54). Ensuring access to healthcare is important in a pandemic scenario.