Evaluating artificial intelligence for comparative radiography.

INTRODUCTION: Comparative radiography is a forensic identification and shortlisting technique based on the comparison of skeletal structures in ante-mortem and post-mortem images. The images (e.g., 2D radiographs or 3D computed tomographies) are manually superimposed and visually compared by a forensic practitioner. It requires a significant amount of time per comparison, limiting its utility in large comparison scenarios. METHODS: We propose and validate a novel framework for automating the shortlisting of candidates using artificial intelligence. It is composed of (1) a segmentation method to delimit skeletal structures' silhouettes in radiographs, (2) a superposition method to generate the best simulated "radiographs" from 3D images according to the segmented radiographs, and (3) a decision-making method for shortlisting all candidates ranked according to a similarity metric. MATERIAL: The dataset is composed of 180 computed tomographies and 180 radiographs where the frontal sinuses are visible. Frontal sinuses are the skeletal structure analyzed due to their high individualization capability. RESULTS: Firstly, we validate two deep learning-based techniques for segmenting the frontal sinuses in radiographs, obtaining high-quality results. Secondly, we study the framework's shortlisting capability using both automatic segmentations and superimpositions. The obtained superimpositions, based only on the superimposition metric, allowed us to filter out 40% of the possible candidates in a completely automatic manner. Thirdly, we perform a reliability study by comparing 180 radiographs against 180 computed tomographies using manual segmentations. The results allowed us to filter out 73% of the possible candidates. Furthermore, the results are robust to inter- and intra-expert-related errors.

Quantifying External Information in Social Network Analysis: An Application to Comparative Mythology.

Social network analysis is a popular tool to understand the relationships between interacting agents by studying the structural properties of their connections. However, this kind of analysis can miss some of the domain-specific knowledge available in the original information domain and its propagation through the associated network. In this work, we develop an extension of classical social network analysis to incorporate external information from the original source of the network. With this extension we propose a new centrality measure, the semantic value, and a new affinity function, the semantic affinity, that establishes fuzzy-like relationships between the different actors in the network. We also propose a new heuristic algorithm based on the shortest capacity problem to compute this new function. As an illustrative case study, we use the novel proposals to analyze and compare the gods and heroes from three different classical mythologies: 1) Greek; 2) Celtic; and 3) Nordic. We study the relationships of each individual mythology and those of the common structure that is formed when we fuse the three of them. We also compare our results with those obtained using other existing centrality measures and embedding approaches. In addition, we test the proposed measures on a classical social network, the Reuters terror news network, as well as in a Twitter network related to the COVID-19 pandemic. We found that the novel method obtains more meaningful comparisons and results than previous existing approaches in every case.

Uptake and Completion of Tuberculosis Preventive Treatment Using 12-Dose, Weekly Isoniazid-Rifapentine Regimen in Bangladesh: A Community-Based Implementation Study.

BACKGROUND: The United Nations high-level meeting (UNHLM) pledged to enroll 30 million in tuberculosis preventive treatment (TPT) by 2022, necessitating TPT expansion to all at tuberculosis (TB) risk. We assessed the uptake and completion of a 12-dose, weekly isoniazid-rifapentine (3HP) TPT regimen. METHODS: Between February 2018 and March 2019 in Dhaka, community-based TPT using 3HP targeted household contacts of 883 confirmed drug-sensitive pulmonary TB patients. Adhering to World Health Organization guidelines, contacts underwent active TB screening before TPT initiation. RESULTS: Of 3193 contacts who were advised health facility visits for screening, 67% (n = 2149) complied. Among these, 1804 (84%) received chest X-rays. Active TB was diagnosed in 39 (2%) contacts; they commenced TB treatment. Over 97% of 1216 contacts began TPT, with completion rates higher among females, those with more education and income, non-slum residents, and those without 3HP-related adverse events. Adverse events, mainly mild, occurred in 5% of participants. CONCLUSIONS: The 3HP regimen, with its short duration, self-administered option, and minimal side effects, achieved satisfactory completion rates. A community-focused TPT approach is feasible, scalable nationally, and aligns with UNHLM targets.

Quantifying geographic heterogeneity in TB incidence and the potential impact of geographically targeted interventions in South and North City Corporations of Dhaka, Bangladesh: a model-based study.

In rapidly growing and high-burden urban centres, identifying tuberculosis (TB) transmission hotspots and understanding the potential impact of interventions can inform future control and prevention strategies. Using data on local demography, TB reports and patient reporting patterns in Dhaka South City Corporation (DSCC) and Dhaka North City Corporation (DNCC), Bangladesh, between 2010 and 2017, we developed maps of TB reporting rates across wards in DSCC and DNCC and identified wards with high rates of reported TB (i.e. 'hotspots') in DSCC and DNCC. We developed ward-level transmission models and estimated the potential epidemiological impact of three TB interventions: active case finding (ACF), mass preventive therapy (PT) and a combination of ACF and PT, implemented either citywide or targeted to high-incidence hotspots. There was substantial geographic heterogeneity in the estimated TB incidence in both DSCC and DNCC: incidence in the highest-incidence wards was over ten times higher than in the lowest-incidence wards in each city corporation. ACF, PT and combined ACF plus PT delivered to 10% of the population reduced TB incidence by a projected 7%-9%, 13%-15% and 19%-23% over five years, respectively. Targeting TB hotspots increased the projected reduction in TB incidence achieved by each intervention 1.4- to 1.8-fold. The geographical pattern of TB notifications suggests high levels of ongoing TB transmission in DSCC and DNCC, with substantial heterogeneity at the ward level. Interventions that reduce transmission are likely to be highly effective and incorporating notification data at the local level can further improve intervention efficiency.

Incorporating patient reporting patterns to evaluate spatially targeted TB interventions.

PURPOSE: Tuberculosis (TB) is geographically heterogeneous, and geographic targeting can improve the impact of TB interventions. However, standard TB notification data may not sufficiently capture this heterogeneity. Better understanding of patient reporting patterns (discrepancies between residence and place of presentation) may improve our ability to use notifications to appropriately target interventions. METHODS: Using demographic data and TB reports from Dhaka North City Corporation and Dhaka South City Corporation, we identified wards of high TB incidence and developed a TB transmission model. We calibrated the model to patient-level data from selected wards under four different reporting pattern assumptions and estimated the relative impact of targeted versus untargeted active case finding. RESULTS: The impact of geographically targeted interventions varied substantially depending on reporting pattern assumptions. The relative reduction in TB incidence, comparing targeted with untargeted active case finding in Dhaka North City Corporation, was 1.20, assuming weak correlation between reporting and residence, versus 2.45, assuming perfect correlation. Similar patterns were observed in Dhaka South City Corporation (1.03 vs. 2.08). CONCLUSIONS: Movement of individuals seeking TB diagnoses may substantially affect ward-level TB transmission. Better understanding of patient reporting patterns can improve estimates of the impact of targeted interventions in reducing TB incidence. Incorporating high-quality patient-level data is critical to optimizing TB interventions.

moGrams: A Network-Based Methodology for Visualizing the Set of Nondominated Solutions in Multiobjective Optimization.

An appropriate visualization of multiobjective nondominated solutions is a valuable asset for decision making. Although there are methods for visualizing the solutions in the design space, they do not provide any information about their relationship. In this paper, we propose a novel methodology that allows the visualization of the nondominated solutions in the design space and their relationships by means of a network. The nodes represent the solutions in the objective space while the edges show the relationships among the solutions in the design space. Our proposal (called moGrams) thus provides a joint visualization of both objective and design spaces. It aims at helping the decision maker to get more understanding of the problem so that (s)he can choose the most appropriate and flexible final solution. moGrams can be applied to any multicriteria problem in which the solutions are related by a similarity metric. Besides, the decision maker interaction is facilitated by modifying the network based on the current preferences to obtain a clearer view. An exhaustive experimental study is performed using four multiobjective problems with a variable number of objectives to show both usefulness and versatility of moGrams. The results exhibit interesting characteristics of our methodology for visualizing and analyzing solutions of multiobjective problems.

Improving maternal health and safety through adherence to postpartum hemorrhage protocol in Latin America.

OBJECTIVE: To determine provider compliance with protocols for the prevention of postpartum hemorrhage and provider characteristics associated with adherence and non-adherence. METHODS: A multicenter descriptive study was conducted involving 78 direct observations of provider-implemented protocols and 52 interviews with Peruvian maternal healthcare providers at 4 Peruvian clinical sites representing the local, regional, and national levels of care. Parturient participants planning a normal vaginal delivery were 17-49 years of age and 34-42 weeks pregnant. Primary outcomes were compared using χ2 testing, while quantitative survey data were evaluated using means, standard deviations, and Student t test or analysis of variance for statistical significance. RESULTS: There were 3 significant differences between the national, regional, and local levels of care: adherence to all 3 interventions (P<0.001); professional experience (P<0.04); and retention of healthcare providers (P<0.001). There were no differences in provider training (P<0.097), and the retention of experienced healthcare providers was not associated with greater adherence to protocols. There were no significant differences in parturient characteristics. CONCLUSION: Individual characteristics and institutional beliefs may have more influence than experience or training on adherence to protocols for prevention of postpartum hemorrhage; addressing these biases may improve patient safety in Peru and throughout Latin America.

Molecular diagnosis of bloodstream infections with a new dual-priming oligonucleotide-based multiplex PCR assay.

Mortality from bloodstream infections (BSIs) correlates with diagnostic delay and the use of inappropriate empirical treatment. Early PCR-based diagnosis could decrease inappropriate treatment, improving patient outcome. The aim of the present study was to assess the clinical utility of this molecular technology to diagnose BSIs. We assessed a new dual-priming oligonucleotide-based multiplex PCR assay, the Magicplex Sepsis Test (MST) (Seegene), along with blood culture (BC). A total of 267 patients from the intensive care unit and haematology and emergency departments were enrolled. Clinical data were also used by physicians to determine the likelihood of infection. Ninety-eight (37 %) specimens were positive: 29 (11 %) by both the MST and BC, 29 (11 %) by the MST only, and 40 (15 %) by BC only. The proportion of agreement between the two methods was 73 % (Cohen's κ: 0.45; 0.28-0.6; indicating fair to moderate agreement). According to clinical assessment, 63 (64 %) positive specimens were considered BSIs: 23 (36 %) were positive by both the MST and BC, 22 (35 %) were positive only by BC, and 18 (29 %) were positive only by the MST. Thirty-eight (14 %) positive specimens by the MST and/or BC were considered as contaminants. Of 101 specimens collected from patients receiving antibiotics, 20 (20 %) were positive by the MST and 32 (32 %) by BC. Sensitivity and specificity were 65 % and 92 %, respectively, for the MST and 71 % and 88 %, respectively for BC. We concluded that the MST shows a high specificity but changes in design are needed to increase bacteraemia detection. For viability in clinical laboratories, technical improvements are also required to further automate the process.