-
World Journal of Emergency Surgery :... Dec 2023To assess the efficacy of artificial intelligence (AI) models in diagnosing and prognosticating acute appendicitis (AA) in adult patients compared to traditional... (Review)
Review
BACKGROUND
To assess the efficacy of artificial intelligence (AI) models in diagnosing and prognosticating acute appendicitis (AA) in adult patients compared to traditional methods. AA is a common cause of emergency department visits and abdominal surgeries. It is typically diagnosed through clinical assessments, laboratory tests, and imaging studies. However, traditional diagnostic methods can be time-consuming and inaccurate. Machine learning models have shown promise in improving diagnostic accuracy and predicting outcomes.
MAIN BODY
A systematic review following the PRISMA guidelines was conducted, searching PubMed, Embase, Scopus, and Web of Science databases. Studies were evaluated for risk of bias using the Prediction Model Risk of Bias Assessment Tool. Data points extracted included model type, input features, validation strategies, and key performance metrics.
RESULTS
In total, 29 studies were analyzed, out of which 21 focused on diagnosis, seven on prognosis, and one on both. Artificial neural networks (ANNs) were the most commonly employed algorithm for diagnosis. Both ANN and logistic regression were also widely used for categorizing types of AA. ANNs showed high performance in most cases, with accuracy rates often exceeding 80% and AUC values peaking at 0.985. The models also demonstrated promising results in predicting postoperative outcomes such as sepsis risk and ICU admission. Risk of bias was identified in a majority of studies, with selection bias and lack of internal validation being the most common issues.
CONCLUSION
AI algorithms demonstrate significant promise in diagnosing and prognosticating AA, often surpassing traditional methods and clinical scores such as the Alvarado scoring system in terms of speed and accuracy.
Topics: Adult; Humans; Artificial Intelligence; Appendicitis; Prognosis; Algorithms; Machine Learning; Acute Disease
PubMed: 38114983
DOI: 10.1186/s13017-023-00527-2 -
Clinical Infectious Diseases : An... Aug 2023Differential time to positivity (DTP), defined as pathogen growth at least 2 hours earlier from catheter versus paired peripheral blood cultures, is sometimes used to... (Meta-Analysis)
Meta-Analysis
BACKGROUND
Differential time to positivity (DTP), defined as pathogen growth at least 2 hours earlier from catheter versus paired peripheral blood cultures, is sometimes used to diagnose central line-associated bloodstream infections (CLABSIs). Previous studies assessing DTP, however, have been small, provided conflicting results, and did not assess heterogeneity across important subgroups.
METHODS
We systematically reviewed the diagnostic characteristics of DTP for CLABSI using MEDLINE, Embase, WoS, CINAHL, LILACS, AMED, and the Cochrane database. Studies were included if they reported sensitivities, specificities, predictive values, likelihood ratios, or 2 × 2 tables of DTP for diagnosing CLABSI. Extracted data were analyzed by using forest plots, bivariate model meta-analysis, and QUADAS-2 quality assessment.
RESULTS
We identified 274 records, of which 23 met the criteria for meta-analysis. Among 2526 suspected CLABSIs, DTP demonstrated a summary sensitivity of 81.3% (95% confidence interval [CI]: 72.8%-87.7%), specificity of 91.8% (95% CI: 84.5%-95.8%), positive likelihood ratio of 9.89 (95% CI: 5.14-19.00), and negative likelihood ratio of 0.20 (95% CI: .14-.30). Covariate analysis based on catheter duration, study design, and patient immune status demonstrated no significant differences. However, DTP performed worse for Staphylococcus aureus (low sensitivity but high specificity) and Candida (high sensitivity but low specificity) compared to other organisms.
CONCLUSIONS
DTP performs well in ruling CLABSIs in or out. Obtaining paired catheter and peripheral blood cultures for DTP when the infectious source is unclear may prevent unnecessary line removal and diagnostic tests. However, this must be balanced against higher contamination rates from catheter cultures.
Topics: Humans; Bacteremia; Time Factors; Catheters; Staphylococcal Infections; Staphylococcus aureus; Sensitivity and Specificity
PubMed: 37062596
DOI: 10.1093/cid/ciad225 -
The American Journal of Gastroenterology May 2020Fecal occult blood tests (FOBTs) are validated only for colorectal cancer (CRC) screening, but are commonly used as a diagnostic test in other clinical settings. We... (Meta-Analysis)
Meta-Analysis
INTRODUCTION
Fecal occult blood tests (FOBTs) are validated only for colorectal cancer (CRC) screening, but are commonly used as a diagnostic test in other clinical settings. We performed a systematic review to assess performance characteristics of FOBT as a diagnostic test for clinical indications.
METHODS
Bibliographic databases were searched to identify studies in adult patients with a specific gastrointestinal symptom or condition who underwent FOBT and a reference test and provided data on diagnoses. Our primary end point was sensitivity. Risk of bias was assessed with the QUADAS-2 tool.
RESULTS
Twenty-two studies met the inclusion criteria: 12 in iron deficiency anemia (IDA) (5 fecal immunochemical (FIT) and 7 guaiac based), 8 in ulcerative colitis (FIT), and 2 in acute diarrhea (guaiac based). Only 2 studies had low risk of bias on all domains of the QUADAS-2. On meta-analysis, FOBT had a sensitivity of 0.58 (95% confidence interval [CI] 0.53-0.63) and a specificity of 0.84 (95% CI 0.75-0.89) in predicting presumptive causes of IDA at endoscopy, with comparable results for guaiac-based tests and FIT. Sensitivity was higher for CRC (0.83) than non-CRC lesions (0.54). FIT had a sensitivity of 0.72 (95% CI 0.57-0.84) and a specificity of 0.80 (95% CI 0.67-0.89) in predicting endoscopic activity in UC. Sensitivities of FOBT for positive stool culture in acute diarrhea were 0.38 and 0.87.
DISCUSSION
Sensitivity of FOBT is poor for IDA: 42% of patients with identifiable causes of IDA had false-negative FOBT. Our results did not show acceptable performance characteristics for FOBT to guide decisions regarding endoscopic evaluation and do not support its use in IDA.
Topics: Anemia, Iron-Deficiency; Colitis, Ulcerative; Colorectal Neoplasms; Culture Techniques; Diarrhea; Early Detection of Cancer; Endoscopy, Digestive System; False Negative Reactions; Feces; Gastrointestinal Hemorrhage; Guaiac; Humans; Immunochemistry; Indicators and Reagents; Occult Blood; Sensitivity and Specificity
PubMed: 31972617
DOI: 10.14309/ajg.0000000000000495 -
Langenbeck's Archives of Surgery Aug 2012Sphincter of Oddi dysfunction (SOD) is a benign pathological syndrome. The clinical manifestations may be a consequence of an anatomical stenosis or sphincter... (Review)
Review
OBJECTIVES
Sphincter of Oddi dysfunction (SOD) is a benign pathological syndrome. The clinical manifestations may be a consequence of an anatomical stenosis or sphincter dysmotility. Manometry is invasive and has an associated morbidity. Non-invasive investigations have been evaluated to ameliorate risk but have unknown efficacy. The review aims to critically appraise current evidence for the diagnosis and management of SOD.
METHODS
A systematic review of articles containing relevant search terms was performed.
RESULTS
Manometry is the current gold standard in selecting which patients are likely to benefit from endoscopic sphincterotomy (ES). It can, however, be misleading. Several non-invasive investigations were identified. These have poor sensitivities and specificities compared to manometry. There is a paucity of data examining the investigation's specific ability to select patients for ES. Outcomes of ES for Type I SOD are favourable irrespective of manometry. Types II and III SOD may respond to an initial trial of medical therapy. Manometry may predict response to ES in Type II SOD, but not in Type III.
CONCLUSIONS
Non-invasive investigations currently lack sufficient sensitivities and specificities for routine use in diagnosing SOD. Type I SOD should be treated with ES without manometry. Manometry may be useful for Type II SOD. However, whilst data is lacking a therapeutic trial of Botox(TM) or trial stenting may bean alternative. Careful and thorough patient counselling is essential. Type III SOD is associated with high complications from manometry and poor outcomes from ES. Alternative diagnoses should be thoroughly sought and its management should be medical.
Topics: Cholangiopancreatography, Endoscopic Retrograde; Diagnostic Imaging; Endosonography; Follow-Up Studies; Humans; Male; Manometry; Prospective Studies; Radionuclide Imaging; Retrospective Studies; Risk Assessment; Severity of Illness Index; Sphincter of Oddi Dysfunction; Sphincterotomy, Endoscopic; Treatment Outcome
PubMed: 22688754
DOI: 10.1007/s00423-012-0971-3 -
Respiration; International Review of... 2017Biopsy-based diagnosis in patients with paraesophageal intrapulmonary tumors suspected of lung cancer is crucial for adequate treatment planning. (Meta-Analysis)
Meta-Analysis Review
BACKGROUND
Biopsy-based diagnosis in patients with paraesophageal intrapulmonary tumors suspected of lung cancer is crucial for adequate treatment planning.
OBJECTIVE
To evaluate the performance of transesophageal endoscopic ultrasound-guided fine-needle aspiration (EUS-FNA) in the diagnosis of intrapulmonary tumors located near or adjacent to the esophagus.
METHODS
We performed a systematic review (PROSPERO, CRD42016033737) and searched MEDLINE, Embase, BIOSIS Previews, and Web of Science on September 22, 2016, without date or language restrictions. We included studies that evaluated the yield and/or sensitivity of EUS-FNA for diagnosing intrapulmonary tumors. Yield was defined as the number of patients in whom EUS-FNA made a biopsy-proven diagnosis (malignant or nonmalignant) relative to the total number of patients on whom EUS-FNA was performed. Sensitivity was defined as the number of patients in whom EUS-FNA made a biopsy-proven diagnosis of malignancy relative to the total number of patients in whom the tumor was found to be malignant. We performed a random-effects meta-analysis.
RESULTS
Of 3,320 search results, 11 studies were included. Ten had a high risk of bias. The total number of patients was 313; the proportion of patients with malignancy ranged from 87 to 100% across these studies. The average yield was 0.90 (95% CI 0.82-0.95) and the average sensitivity was 0.92 (0.83-0.96). In the subgroup of prospective studies (n = 3), the average yield was 0.80 (0.56-0.93) and the average sensitivity was 0.83 (0.58-0.95). EUS-FNA-induced complications were reported for 5/256 patients (2.0%) for whom this information was available.
CONCLUSIONS
Although the number of high-quality studies is limited, these findings suggest that EUS-FNA is safe and has a high yield for diagnosing intrapulmonary tumors.
Topics: Endoscopic Ultrasound-Guided Fine Needle Aspiration; Esophagoscopy; Humans; Lung Neoplasms; Sensitivity and Specificity
PubMed: 27926910
DOI: 10.1159/000452958 -
World Journal of Surgical Oncology Feb 2024The application of machine learning (ML) for identifying early gastric cancer (EGC) has drawn increasing attention. However, there lacks evidence-based support for its... (Meta-Analysis)
Meta-Analysis Review
BACKGROUND
The application of machine learning (ML) for identifying early gastric cancer (EGC) has drawn increasing attention. However, there lacks evidence-based support for its specific diagnostic performance. Hence, this systematic review and meta-analysis was implemented to assess the performance of image-based ML in EGC diagnosis.
METHODS
We performed a comprehensive electronic search in PubMed, Embase, Cochrane Library, and Web of Science up to September 25, 2022. QUADAS-2 was selected to judge the risk of bias of included articles. We did the meta-analysis using a bivariant mixed-effect model. Sensitivity analysis and heterogeneity test were performed.
RESULTS
Twenty-one articles were enrolled. The sensitivity (SEN), specificity (SPE), and SROC of ML-based models were 0.91 (95% CI: 0.87-0.94), 0.85 (95% CI: 0.81-0.89), and 0.94 (95% CI: 0.39-1.00) in the training set and 0.90 (95% CI: 0.86-0.93), 0.90 (95% CI: 0.86-0.92), and 0.96 (95% CI: 0.19-1.00) in the validation set. The SEN, SPE, and SROC of EGC diagnosis by non-specialist clinicians were 0.64 (95% CI: 0.56-0.71), 0.84 (95% CI: 0.77-0.89), and 0.80 (95% CI: 0.29-0.97), and those by specialist clinicians were 0.80 (95% CI: 0.74-0.85), 0.88 (95% CI: 0.85-0.91), and 0.91 (95% CI: 0.37-0.99). With the assistance of ML models, the SEN of non-specialist physicians in the diagnosis of EGC was significantly improved (0.76 vs 0.64).
CONCLUSION
ML-based diagnostic models have greater performance in the identification of EGC. The diagnostic accuracy of non-specialist clinicians can be improved to the level of the specialists with the assistance of ML models. The results suggest that ML models can better assist less experienced clinicians in diagnosing EGC under endoscopy and have broad clinical application value.
Topics: Humans; Stomach Neoplasms; Endoscopy; Machine Learning
PubMed: 38297303
DOI: 10.1186/s12957-024-03321-9 -
Clinics (Sao Paulo, Brazil) 2024Autism Spectrum Disorder (ASD) is a heterogeneous neurodevelopmental disorder, with main manifestations related to communication, social interaction, and behavioral... (Meta-Analysis)
Meta-Analysis Review
INTRODUCTION
Autism Spectrum Disorder (ASD) is a heterogeneous neurodevelopmental disorder, with main manifestations related to communication, social interaction, and behavioral patterns. The slight dynamics of change in the child over time require that the onset of clinical manifestations presented by the child be more valued, with the aim of stabilizing the condition. Faced with a variety of methods for diagnosing ASD, the question arises as to which method should be used. This systematic review aims to recommend the best tools to perform screening and diagnosis.
METHODOLOGY
This systematic review followed the PRISMA guidelines. The databases MEDLINE, Embase, CENTRAL (Cochrane), and Lilacs were accessed, and gray and manual searches were performed. The search strategy was created with terms referring to autism and the diagnosis/broad filter. The studies were qualitatively evaluated and quantitatively. Statistical analysis was performed using Meta-diSc-2.0 software, the confidence interval was 95 %.
RESULTS
The M-CHAT-R/F tool demonstrated a sensitivity of 78 % (95 % CI 0.57‒0.91) and specificity of 0.98 (95 % CI 0.88-1.00). The diagnostic tools demonstrated sensitivity and specificity respectively of: ADOS, sensitivity of 87 % (95 % CI 0.79‒0.92) and specificity 75 % (95 % CI 0.73‒0.78); ADI-R demonstrated test sensitivity of 77 % (95 % CI 0.56‒0.90) and specificity 68 % (95 % CI 0.52‒0.81), CARS test sensitivity was 89 % (95 % CI 0.78‒0.95) and specificity 79 % (95 % CI 0.65‒0.88).
CONCLUSION
It is mandatory to apply a screening test, the most recommended being the M-CHAT-R/F. For diagnosis CARS and ADOS are the most recommended tools.
Topics: Child; Humans; Autism Spectrum Disorder; Sensitivity and Specificity; Mass Screening; Communication; Research Design
PubMed: 38484581
DOI: 10.1016/j.clinsp.2023.100323 -
Annals of Noninvasive Electrocardiology... Oct 2012Coronary artery disease (CAD) has a significant disease burden making early diagnosis and management imperative. Magnetocardiography (MCG) is a relatively new... (Meta-Analysis)
Meta-Analysis Review
BACKGROUND
Coronary artery disease (CAD) has a significant disease burden making early diagnosis and management imperative. Magnetocardiography (MCG) is a relatively new noninvasive technique that allows diagnosis of CAD by recording the magnetic fields generated by the electrical activity of the heart.
METHODS
We searched MEDLINE and the Cochrane Central Register of Controlled Trials for prospective studies that evaluated the test characteristics (e.g., sensitivity, specificity, likelihood ratios) of MCG for detection of CAD. Studies were included if they evaluated either patients with stable CAD documented by angiogram or patients presenting initially with acute coronary syndrome and subsequently diagnosed with CAD. The quality of included studies was assessed using an adaptation of the Quality Assessment of Diagnostic Accuracy Studies (QUADAS) tool. We performed meta-analyses of sensitivity, specificity and positive and negative likelihood ratios using Meta-DiSc software.
RESULTS
Screening of titles and abstracts followed by full-text review yielded seven studies that met our inclusion criteria. Meta-analyses yielded a pooled sensitivity of 83% (95% confidence interval [CI] 80% to 86%) and a specificity of 77% (95% CI 73% to 81%). The pooled positive likelihood ratio was 3.92 (95% CI 2.30 to 6.66) and negative likelihood ratio was 0.20 (95% CI 0.12 to 0.35). Significant heterogeneity was present in all meta-analyses.
CONCLUSIONS
The pooled test characteristics for MCG are similar to those of existing noninvasive modalities for diagnosing CAD. Our results suggest that MCG is a potential complementary or alternative tool for noninvasive detection of CAD.
Topics: Coronary Artery Disease; Humans; Magnetocardiography; Predictive Value of Tests; Prospective Studies; Sensitivity and Specificity
PubMed: 23094875
DOI: 10.1111/j.1542-474X.2012.00538.x -
Journal of Affective Disorders Apr 2018Reliability of schizoaffective disorder (SAD) diagnoses is low in adults but unclear in children and adolescents (CAD). We estimate the test-retest reliability of SAD... (Meta-Analysis)
Meta-Analysis Review
OBJECTIVES
Reliability of schizoaffective disorder (SAD) diagnoses is low in adults but unclear in children and adolescents (CAD). We estimate the test-retest reliability of SAD and its key differential diagnoses (schizophrenia, bipolar disorder, and unipolar depression).
METHODS
Systematic literature search of Medline, Embase, and PsycInfo for studies on test-retest reliability of SAD, in CAD. Cohen's kappa was extracted from studies. We performed meta-analysis for kappa, including subgroup and sensitivity analysis (PROSPERO protocol: CRD42013006713).
RESULTS
Out of > 4000 records screened, seven studies were included. We estimated kappa values of 0.27 [95%-CI: 0.07 0.47] for SAD, 0.56 [0.29; 0.83] for schizophrenia, 0.64 [0.55; 0.74] for bipolar disorder, and 0.66 [0.52; 0.81] for unipolar depression. In 5/7 studies kappa of SAD was lower than that of schizophrenia; similar trends emerged for bipolar disorder (4/5) and unipolar depression (2/3). Estimates of positive agreement of SAD diagnoses supported these results.
LIMITATIONS
The number of studies and patients included is low.
CONCLUSIONS
The point-estimate of the test-retest reliability of schizoaffective disorder is only fair, and lower than that of its main differential diagnoses. All kappa values under study were lower in children and adolescents samples than those reported for adults. Clinically, schizoaffective disorder should be diagnosed in strict adherence to the operationalized criteria and ought to be re-evaluated regularly. Should larger studies confirm the insufficient reliability of schizoaffective disorder in children and adolescents, the clinical value of the diagnosis is highly doubtful.
Topics: Adolescent; Adult; Bipolar Disorder; Child; Depressive Disorder, Major; Diagnosis, Differential; Female; Humans; Male; Psychotic Disorders; Reproducibility of Results; Schizophrenia; Young Adult
PubMed: 29360577
DOI: 10.1016/j.jad.2017.12.070 -
Medicina (Kaunas, Lithuania) May 2023The development of liver fibrosis as a consequence of continuous inflammation represents a turning point in the evolution of chronic liver diseases. The recent... (Review)
Review
The development of liver fibrosis as a consequence of continuous inflammation represents a turning point in the evolution of chronic liver diseases. The recent developments of artificial intelligence (AI) applications show a high potential for improving the accuracy of diagnosis, involving large sets of clinical data. For this reason, the aim of this systematic review is to provide a comprehensive overview of current AI applications and analyze the accuracy of these systems to perform an automated diagnosis of liver fibrosis. We searched PubMed, Cochrane Library, EMBASE, and WILEY databases using predefined keywords. Articles were screened for relevant publications about AI applications capable of diagnosing liver fibrosis. Exclusion criteria were animal studies, case reports, abstracts, letters to the editor, conference presentations, pediatric studies, studies written in languages other than English, and editorials. Our search identified a total of 24 articles analyzing the automated imagistic diagnosis of liver fibrosis, out of which six studies analyze liver ultrasound images, seven studies analyze computer tomography images, five studies analyze magnetic resonance images, and six studies analyze liver biopsies. The studies included in our systematic review showed that AI-assisted non-invasive techniques performed as accurately as human experts in detecting and staging liver fibrosis. Nevertheless, the findings of these studies need to be confirmed through clinical trials to be implemented into clinical practice. The current systematic review provides a comprehensive analysis of the performance of AI systems in diagnosing liver fibrosis. Automatic diagnosis, staging, and risk stratification for liver fibrosis is currently possible considering the accuracy of the AI systems, which can overcome the limitations of non-invasive diagnosis methods.
Topics: Animals; Humans; Child; Artificial Intelligence; Liver Cirrhosis; Biopsy; Databases, Factual; Inflammation
PubMed: 37241224
DOI: 10.3390/medicina59050992