Search

Objective: This study aimed to summarize the characteristics and methodological quality of systematic reviews on the application of artificial intelligence (AI) in clinical diagnosis and treatment. Methods: We systematically searched seven English- and Chinese-language literature databases to identify systematic reviews on the application of AI, deep learning, or machine learning in the diagnosis and treatment of any disease published in 2020. We evaluated the methodological quality of the included systematic reviews using “A Measurement tool for the assessment of multiple systematic reviews” (AMSTAR). We also conducted meta-analyses on the diagnostic accuracy of AI on selected disease categories with a large number of included studies and low clinical heterogeneity. Results: A total of 40 systematic reviews reporting 1,083 original studies were included, covering 31 diseases from 11 groups of diseases. Eleven systematic reviews were related to neoplasms and nine were systematic reviews related to diseases of the digestive system. We selected digestive system diseases for the meta-analysis. The pooled sensitivities (with 95% confidence interval (CI)) of AI to assist the diagnosis of helicobacter pylori, gastrointestinal ulcers, hemorrhage, esophageal tumors, gastric tumors, and intestinal tumors (with 95% CI) were 0.91 (0.83–0.95), 0.99 (0.76–1.00), 0.95 (0.83–0.99), 0.90 (0.85–0.93), 0.90 (0.82–0.95), and 0.93 (0.88–0.96), respectively, and the pooled specificities were 0.82 (0.77–0.87), 0.97 (0.86–1.00), 1.00 (0.99–1.00), 0.80 (0.71–0.87), 0.93 (0.87–0.97), and 0.89 (0.85–0.92), respectively. The AMSTAR items “the list of included studies” (n = 39, 97.5%) and “the characteristics of the included studies” (n = 39, 97.5%) had the highest compliance among the reviews; the compliance was relatively low to the items “the consideration of publication status” (n = 1, 2.5%), “the consideration of scientific quality” (n = 19, 47.5%), “data synthesis methods” (n = 18, 45.0%), and “ the evaluation of publication bias” (n = 13, 32.5%). Conclusions: The main subjects of systematic reviews on AI applications in clinical diagnosis and treatment published in 2020 were diseases of the digestive system and neoplasms. The methodological quality of the systematic reviews on AI needs to be improved, paying particular attention to publication bias and the rigorous evaluation of the quality of the included studies.

Objective Complete and transparent reporting is of critical importance for randomized controlled trials (RCTs). The present study aimed to determine the reporting quality and methodological quality of RCTs for interventions involving artificial intelligence (AI) and their protocols.Methods We searched MEDLINE (via PubMed), Embase, Web of Science, CBMdisc, Wanfang Data, and CNKI from January 1, 2016, to November 11, 2020, to collect RCTs involving AI. We also extracted the protocol of each included RCT if it could be obtained. CONSORT-AI (Consolidated Standards of Reporting Trials–Artificial Intelligence) statement and Cochrane Collaboration's tool for assessing risk of bias (ROB) were used to evaluate the reporting quality and methodological quality, respectively, and SPIRIT-AI (The Standard Protocol Items: Recommendations for Interventional Trials–Artificial Intelligence) statement was used to evaluate the reporting quality of the protocols. The associations of the reporting rate of CONSORT-AI with the publication year, journal's impact factor (IF), number of authors, sample size, and first author's country were analyzed univariately using Pearson's chi-squared test, or Fisher's exact test if the expected values in any of the cells were below 5. The compliance of the retrieved protocols to SPIRIT-AI was presented descriptively.Results Overall, 29 RCTs and three protocols were considered eligible. The CONSORT-AI items “title and abstract” and “interpretation of results” were reported by all RCTs, with the items with the lowest reporting rates being “funding” (0), “implementation” (3.5%), and “harms” (3.5%). The risk of bias was high in 13 (44.8%) RCTs and not clear in 15 (51.7%) RCTs. Only one RCT (3.5%) had a low risk of bias. The compliance was not significantly different in terms of the publication year, journal's IF, number of authors, sample size, or first author's country. Ten of the 35 SPIRIT-AI items (funding, participant timeline, allocation concealment mechanism, implementation, data management, auditing, declaration of interests, access to data, informed consent materials and biological specimens) were not reported by any of the three protocols.Conclusions The reporting and methodological quality of RCTs involving AI need to be improved. Because of the limited availability of protocols, their quality could not be fully judged. Following the CONSORT-AI and SPIRIT-AI statements and with appropriate guidance on the risk of bias when designing and reporting AI-related RCTs can promote standardization and transparency.

Loss to follow-up (LTFU) is a common problem in many epidemiological studies. In antiretroviral treatment (ART) programs for patients with human immunodeficiency virus (HIV), mortality estimates can be biased if the LTFU mechanism is non-ignorable, that is, mortality differs between lost and retained patients. In this setting, routine procedures for handling missing data may lead to biased estimates. To appropriately deal with non-ignorable LTFU, explicit modeling of the missing data mechanism is needed. This can be based on additional outcome ascertainment for a sample of patients LTFU, for example, through linkage to national registries or through survey-based methods. In this paper, we demonstrate how this additional information can be used to construct estimators based on inverse probability weights (IPW) or multiple imputation. We use simulations to contrast the performance of the proposed estimators with methods widely used in HIV cohort research for dealing with missing data. The practical implications of our approach are illustrated using South African ART data, which are partially linkable to South African national vital registration data. Our results demonstrate that while IPWs and proper imputation procedures can be easily constructed from additional outcome ascertainment to obtain valid overall estimates, neglecting non-ignorable LTFU can result in substantial bias. We believe the proposed estimators are readily applicable to a growing number of studies where LTFU is appreciable, but additional outcome data are available through linkage or surveys of patients LTFU. Copyright © 2013 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]

This study describes the first detailed linkage maps of two bermudagrass species, Cynodon dactylon (T89) and Cynodon transvaalensis (T574), based on single-dose restriction fragments (SDRFs). The mapping population consisted of 113 F1 progeny of a cross between the two parents. Loci were generated using 179 bermudagrass genomic clones and 50 heterologous cDNAs from Pennisetum and rice. The map of T89 is based on 155 SDRFs and 17 double-dose restriction fragments on 35 linkage groups, with an average marker spacing of 15.3 cM. The map of T574 is based on 77 SDRF loci on 18 linkage groups with an average marker spacing of 16.5 cM. About 16 T89 linkage groups were arranged into four complete and eight into four incomplete homologous sets, while 15 T574 linkage groups were arranged into seven complete homologous sets, all on the basis of multi-locus probes and repulsion linkages. Eleven T89 and three T574 linkage groups remain unassigned. In each parent consensus maps were built based on alignments of homologous linkage groups. Four ancestral chromosomes were inferred after aligning T89 and T574 parental consensus maps using multi-locus probes. The inferred ancestral marker orders were used in comparisons to a detailed Sorghum linkage map using 40 common probes, and to the rice genome sequence using 98 significant BLAST hits, to find regions of colinearity. Using these maps we have estimated the recombinational length of the T89 and T574 genomes at 3,012 and 1,569 cM, respectively, which are 61 and 62% covered by our maps. [ABSTRACT FROM AUTHOR]

Abstract Background Genetically anchored physical maps of large eukaryotic genomes have proven useful both for their intrinsic merit and as an adjunct to genome sequencing. Cultivated tetraploid cottons, Gossypium hirsutum and G. barbadense, share a common ancestor formed by a merger of the A and D genomes about 1-2 million years ago. Toward the long-term goal of characterizing the spectrum of diversity among cotton genomes, the worldwide cotton community has prioritized the D genome progenitor Gossypium raimondii for complete sequencing. Results A whole genome physical map of G. raimondii, the putative D genome ancestral species of tetraploid cottons was assembled, integrating genetically-anchored overgo hybridization probes, agarose based fingerprints and 'high information content fingerprinting' (HICF). A total of 13,662 BAC-end sequences and 2,828 DNA probes were used in genetically anchoring 1585 contigs to a cotton consensus genetic map, and 370 and 438 contigs, respectively to Arabidopsis thaliana (AT) and Vitis vinifera (VV) whole genome sequences. Conclusion Several lines of evidence suggest that the G. raimondii genome is comprised of two qualitatively different components. Much of the gene rich component is aligned to the Arabidopsis and Vitis vinifera genomes and shows promise for utilizing translational genomic approaches in understanding this important genome and its resident genes. The integrated genetic-physical map is of value both in assembling and validating a planned reference sequence.

Abstract Background High quality annotation of the genes and transposable elements in complex genomes requires a human-curated integration of multiple sources of computational evidence. These evidences include results from a diversity of ab initio prediction programs as well as homology-based searches. Most of these programs operate on a single contiguous sequence at a time, and the results are generated in a diverse array of readable formats that must be translated to a standardized file format. These translated results must then be concatenated into a single source, and then presented in an integrated form for human curation. Results We have designed, implemented, and assessed a Perl-based workflow named DAWGPAWS for the generation of computational results for human curation of the genes and transposable elements in plant genomes. The use of DAWGPAWS was found to accelerate annotation of 80–200 kb wheat DNA inserts in bacterial artificial chromosome (BAC) vectors by approximately twenty-fold and to also significantly improve the quality of the annotation in terms of completeness and accuracy. Conclusion The DAWGPAWS genome annotation pipeline fills an important need in the annotation of plant genomes by generating computational evidences in a high throughput manner, translating these results to a common file format, and facilitating the human curation of these computational results. We have verified the value of DAWGPAWS by using this pipeline to annotate the genes and transposable elements in 220 BAC insertions from the hexaploid wheat genome (Triticum aestivum L.). DAWGPAWS can be applied to annotation efforts in other plant genomes with minor modifications of program-specific configuration files, and the modular design of the workflow facilitates integration into existing pipelines.

Abstract Background Many plant genomes are resistant to whole-genome assembly due to an abundance of repetitive sequence, leading to the development of gene-rich sequencing techniques. Two such techniques are hypomethylated partial restriction (HMPR) and methylation spanning linker libraries (MSLL). These libraries differ from other gene-rich datasets in having larger insert sizes, and the MSLL clones are designed to provide reads localized to "epigenetic boundaries" where methylation begins or ends. Results A large-scale study in maize generated 40,299 HMPR sequences and 80,723 MSLL sequences, including MSLL clones exceeding 100 kb. The paired end reads of MSLL and HMPR clones were shown to be effective in linking existing gene-rich sequences into scaffolds. In addition, it was shown that the MSLL clones can be used for anchoring these scaffolds to a BAC-based physical map. The MSLL end reads effectively identified epigenetic boundaries, as indicated by their preferential alignment to regions upstream and downstream from annotated genes. The ability to precisely map long stretches of fully methylated DNA sequence is a unique outcome of MSLL analysis, and was also shown to provide evidence for errors in gene identification. MSLL clones were observed to be significantly more repeat-rich in their interiors than in their end reads, confirming the correlation between methylation and retroelement content. Both MSLL and HMPR reads were found to be substantially gene-enriched, with the SalI MSLL libraries being the most highly enriched (31% align to an EST contig), while the HMPR clones exhibited exceptional depletion of repetitive DNA (to ~11%). These two techniques were compared with other gene-enrichment methods, and shown to be complementary. Conclusion MSLL technology provides an unparalleled approach for mapping the epigenetic status of repetitive blocks and for identifying sequences mis-identified as genes. Although the types and natures of epigenetic boundaries are barely understood at this time, MSLL technology flags both approximate boundaries and methylated genes that deserve additional investigation. MSLL and HMPR sequences provide a valuable resource for maize genome annotation, and are a uniquely valuable complement to any plant genome sequencing project. In order to make these results fully accessible to the community, a web display was developed that shows the alignment of MSLL, HMPR, and other gene-rich sequences to the BACs; this display is continually updated with the latest ESTs and BAC sequences.

Introduction: A growing number of paediatric core outcome sets (COS) have been developed in the past 20 years. Previous studies have provided many useful insights into the uptake of COS. In addition to the awareness of COS among clinical trialists, the COS development process (especially patient participation) and the actions of the developers can promote COS uptake. However, the uptake of COS in paediatric clinical trials needs to be further explored. The aim of this study is to provide information on the rationale and use of paediatric COS in clinical trials, and to analyse in depth the awareness and views of COS developers and clinical trialists about the development and use of COS., Methods and Analysis: We will include all paediatric COS identified in our previous systematic review and those subsequently included in the Core Outcome Measures in Effectiveness Trials (COMET) database. We will extract the target condition, population, intervention, list of core outcomes and the details of patient involvement. Next, we will search the Clinicaltrials.gov and WHO International Clinical Trials Registry Platform for trials on health conditions addressed by the identified COS. We will assess the comparability of the scopes in each COS-trial pair and determine for the outcomes in each clinical trial if they match exactly or generally, or if they do not match, with the outcomes of their respective COS. Finally, we will conduct a survey and semistructured interviews among COS developers and clinical trialists to examine their views., Ethics and Dissemination: Ethical approval for the study has been granted by the ethics committee of the Institute of Health Data Science, Lanzhou University (No. HDS-202405-01). This study was registered on COMET (https://www.comet-initiative.org/Studies/Details/3122)., Competing Interests: Competing interests: No, there are no competing interests., (© Author(s) (or their employer(s)) 2025. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ Group.)