Journal Article > Case Report/SeriesFull Text
ACG Case Rep J. 2020 December 1; Volume 7 (Issue 12); e00495.; DOI:10.14309/crj.0000000000000495
Abulaimoun S, Abushalha K, Reddymasu S, Teruya B, Natarajan N
ACG Case Rep J. 2020 December 1; Volume 7 (Issue 12); e00495.; DOI:10.14309/crj.0000000000000495
Hemophagocytic lymphohistiocytosis is a syndrome characterized by excessive immune activation. Timely diagnosis can be challenging, and prompt treatment is the only hope for survival. We present an adult patient with a history of alcohol dependence, who presented with fatigue, bilateral lower extremity edema, and orange-colored urine. Clinical workup revealed abnormal liver function tests, elevated ferritin, cytopenia, and lymphadenopathy. Eventually, he was diagnosed with hemophagocytic lymphohistiocytosis. This case report encourages gastroenterologists to maintain a high index of suspicion when a patient presents with liver failure, hyperferritinemia, and cytopenia because they may be the first healthcare professionals to evaluate these patients.
Journal Article > ResearchFull Text
ACG Case Rep J. 2023 September 15; Volume 25; e39736.; DOI:10.2196/39736
Orel EB, Ciglenecki I, Thiabaud A, Temerev A, Calmy A, et al.
ACG Case Rep J. 2023 September 15; Volume 25; e39736.; DOI:10.2196/39736
BACKGROUND
Literature reviews (LRs) identify, evaluate, and synthesize relevant papers to a particular research question to advance understanding and support decision-making. However, LRs, especially traditional systematic reviews, are slow, resource-intensive, and become outdated quickly.
OBJECTIVE
LiteRev is an advanced and enhanced version of an existing automation tool designed to assist researchers in conducting LRs through the implementation of cutting-edge technologies such as natural language processing and machine learning techniques. In this paper, we present a comprehensive explanation of LiteRev’s capabilities, its methodology, and an evaluation of its accuracy and efficiency to a manual LR, highlighting the benefits of using LiteRev.
METHODS
Based on the user’s query, LiteRev performs an automated search on a wide range of open-access databases and retrieves relevant metadata on the resulting papers, including abstracts or full texts when available. These abstracts (or full texts) are text processed and represented as a term frequency-inverse document frequency matrix. Using dimensionality reduction (pairwise controlled manifold approximation) and clustering (hierarchical density-based spatial clustering of applications with noise) techniques, the corpus is divided into different topics described by a list of the most important keywords. The user can then select one or several topics of interest, enter additional keywords to refine its search, or provide key papers to the research question. Based on these inputs, LiteRev performs a k-nearest neighbor (k-NN) search and suggests a list of potentially interesting papers. By tagging the relevant ones, the user triggers new k-NN searches until no additional paper is suggested for screening. To assess the performance of LiteRev, we ran it in parallel to a manual LR on the burden and care for acute and early HIV infection in sub-Saharan Africa. We assessed the performance of LiteRev using true and false predictive values, recall, and work saved over sampling.
RESULTS
LiteRev extracted, processed, and transformed text into a term frequency-inverse document frequency matrix of 631 unique papers from PubMed. The topic modeling module identified 16 topics and highlighted 2 topics of interest to the research question. Based on 18 key papers, the k-NNs module suggested 193 papers for screening out of 613 papers in total (31.5% of the whole corpus) and correctly identified 64 relevant papers out of the 87 papers found by the manual abstract screening (recall rate of 73.6%). Compared to the manual full text screening, LiteRev identified 42 relevant papers out of the 48 papers found manually (recall rate of 87.5%). This represents a total work saved over sampling of 56%.
CONCLUSIONS
We presented the features and functionalities of LiteRev, an automation tool that uses natural language processing and machine learning methods to streamline and accelerate LRs and support researchers in getting quick and in-depth overviews on any topic of interest.
Literature reviews (LRs) identify, evaluate, and synthesize relevant papers to a particular research question to advance understanding and support decision-making. However, LRs, especially traditional systematic reviews, are slow, resource-intensive, and become outdated quickly.
OBJECTIVE
LiteRev is an advanced and enhanced version of an existing automation tool designed to assist researchers in conducting LRs through the implementation of cutting-edge technologies such as natural language processing and machine learning techniques. In this paper, we present a comprehensive explanation of LiteRev’s capabilities, its methodology, and an evaluation of its accuracy and efficiency to a manual LR, highlighting the benefits of using LiteRev.
METHODS
Based on the user’s query, LiteRev performs an automated search on a wide range of open-access databases and retrieves relevant metadata on the resulting papers, including abstracts or full texts when available. These abstracts (or full texts) are text processed and represented as a term frequency-inverse document frequency matrix. Using dimensionality reduction (pairwise controlled manifold approximation) and clustering (hierarchical density-based spatial clustering of applications with noise) techniques, the corpus is divided into different topics described by a list of the most important keywords. The user can then select one or several topics of interest, enter additional keywords to refine its search, or provide key papers to the research question. Based on these inputs, LiteRev performs a k-nearest neighbor (k-NN) search and suggests a list of potentially interesting papers. By tagging the relevant ones, the user triggers new k-NN searches until no additional paper is suggested for screening. To assess the performance of LiteRev, we ran it in parallel to a manual LR on the burden and care for acute and early HIV infection in sub-Saharan Africa. We assessed the performance of LiteRev using true and false predictive values, recall, and work saved over sampling.
RESULTS
LiteRev extracted, processed, and transformed text into a term frequency-inverse document frequency matrix of 631 unique papers from PubMed. The topic modeling module identified 16 topics and highlighted 2 topics of interest to the research question. Based on 18 key papers, the k-NNs module suggested 193 papers for screening out of 613 papers in total (31.5% of the whole corpus) and correctly identified 64 relevant papers out of the 87 papers found by the manual abstract screening (recall rate of 73.6%). Compared to the manual full text screening, LiteRev identified 42 relevant papers out of the 48 papers found manually (recall rate of 87.5%). This represents a total work saved over sampling of 56%.
CONCLUSIONS
We presented the features and functionalities of LiteRev, an automation tool that uses natural language processing and machine learning methods to streamline and accelerate LRs and support researchers in getting quick and in-depth overviews on any topic of interest.
Journal Article > ResearchFull Text
ACG Case Rep J. 2024 March 21; Volume 70; 102527.; DOI:10.1016/j.eclinm.2024.102527
Wobudeya E, Nanfuka M, Ton Nu Nguyet MH, Taguebue JV, Moh R, et al.
ACG Case Rep J. 2024 March 21; Volume 70; 102527.; DOI:10.1016/j.eclinm.2024.102527
BACKGROUND
Childhood tuberculosis (TB) remains underdiagnosed largely because of limited awareness and poor access to all or any of specimen collection, molecular testing, clinical evaluation, and chest radiography at low levels of care. Decentralising childhood TB diagnostics to district hospitals (DH) and primary health centres (PHC) could improve case detection.
METHODS
We conducted an operational research study using a pre-post intervention cross-sectional study design in 12 DHs and 47 PHCs of 12 districts across Cambodia, Cameroon, Côte d'Ivoire, Mozambique, Sierra Leone and Uganda. The intervention included 1) a comprehensive diagnosis package at patient-level with tuberculosis screening for all sick children and young adolescents <15 years, and clinical evaluation, Xpert Ultra-testing on respiratory and stool samples, and chest radiography for children with presumptive TB, and 2) two decentralisation approaches (PHC-focused or DH-focused) to which districts were randomly allocated at country level. We collected aggregated and individual data. We compared the proportion of tuberculosis detection in children and young adolescents <15 years pre-intervention (01 August 2018-30 November 2019) versus during intervention (07 March 2020-30 September 2021), overall and by decentralisation approach. This study is registered with ClinicalTrials.gov, NCT04038632.
FINDINGS
TB was diagnosed in 217/255,512 (0.08%) children and young adolescent <15 years attending care pre-intervention versus 411/179,581 (0.23%) during intervention, (OR: 3.59 [95% CI 1.99-6.46], p-value<0.0001; p-value = 0.055 after correcting for over-dispersion). In DH-focused districts, TB diagnosis was 80/122,570 (0.07%) versus 302/86,186 (0.35%) (OR: 4.07 [1.86-8.90]; p-value = 0.0005; p-value = 0.12 after correcting for over-dispersion); and 137/132,942 (0.10%) versus 109/93,395 (0.11%) in PHC-focused districts, respectively (OR: 2.92 [1.25-6.81; p-value = 0.013; p-value = 0.26 after correcting for over-dispersion).
INTERPRETATION
Decentralising and strengthening childhood TB diagnosis at lower levels of care increases tuberculosis case detection but the difference was not statistically significant.
Childhood tuberculosis (TB) remains underdiagnosed largely because of limited awareness and poor access to all or any of specimen collection, molecular testing, clinical evaluation, and chest radiography at low levels of care. Decentralising childhood TB diagnostics to district hospitals (DH) and primary health centres (PHC) could improve case detection.
METHODS
We conducted an operational research study using a pre-post intervention cross-sectional study design in 12 DHs and 47 PHCs of 12 districts across Cambodia, Cameroon, Côte d'Ivoire, Mozambique, Sierra Leone and Uganda. The intervention included 1) a comprehensive diagnosis package at patient-level with tuberculosis screening for all sick children and young adolescents <15 years, and clinical evaluation, Xpert Ultra-testing on respiratory and stool samples, and chest radiography for children with presumptive TB, and 2) two decentralisation approaches (PHC-focused or DH-focused) to which districts were randomly allocated at country level. We collected aggregated and individual data. We compared the proportion of tuberculosis detection in children and young adolescents <15 years pre-intervention (01 August 2018-30 November 2019) versus during intervention (07 March 2020-30 September 2021), overall and by decentralisation approach. This study is registered with ClinicalTrials.gov, NCT04038632.
FINDINGS
TB was diagnosed in 217/255,512 (0.08%) children and young adolescent <15 years attending care pre-intervention versus 411/179,581 (0.23%) during intervention, (OR: 3.59 [95% CI 1.99-6.46], p-value<0.0001; p-value = 0.055 after correcting for over-dispersion). In DH-focused districts, TB diagnosis was 80/122,570 (0.07%) versus 302/86,186 (0.35%) (OR: 4.07 [1.86-8.90]; p-value = 0.0005; p-value = 0.12 after correcting for over-dispersion); and 137/132,942 (0.10%) versus 109/93,395 (0.11%) in PHC-focused districts, respectively (OR: 2.92 [1.25-6.81; p-value = 0.013; p-value = 0.26 after correcting for over-dispersion).
INTERPRETATION
Decentralising and strengthening childhood TB diagnosis at lower levels of care increases tuberculosis case detection but the difference was not statistically significant.