Meta-analysis for diagnostic tests

Artificial intelligence as diagnostic modality for keratoconus: A systematic review and meta-analysis

Objectives

The challenges in diagnosing keratoconus (KC) have led researchers to explore the use of artificial intelligence (AI) as a diagnostic tool. AI has emerged as a new way to improve the efficiency of KC diagnosis. This study analyzed the use of AI as a diagnostic modality for KC.

Methods

This study used a systematic review and meta-analysis following the 2020 Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. We searched selected databases using a combination of search terms: "((Artificial Intelligence) OR (Diagnostic Modality)) AND (Keratoconus)" from PubMed, Medline, and ScienceDirect within the last 5 years (2018-2023). Following a systematic review protocol, we selected 11 articles and 6 articles were eligible for final analysis. The relevant data were analyzed with Review Manager 5.4 software and the final output was presented in a forest plot.

Results

This research found neural networks as the most used AI model in diagnosing KC. Neural networks and naïve bayes showed the highest accuracy of AI in diagnosing KC with a sensitivity of 1.00, while random forests were >0.90. All studies in each group have proven high sensitivity and specificity over 0.90.

Conclusions

AI potentially makes a better diagnosis of the KC with its high performance, particularly on sensitivity and specificity, which can help clinicians make medical decisions about an individual patient.

Systematic review of early warning signs of relapse and behavioural antecedents of symptom worsening in people living with schizophrenia spectrum disorders

BACKGROUND

Identification of the early warning signs (EWS) of relapse is key to relapse prevention in schizophrenia spectrum disorders, however, limitations to their precision have been reported. Substantial methodological innovations have recently been applied to the prediction of psychotic relapse and to individual psychotic symptoms. However, there has been no systematic review that has integrated findings across these two related outcomes and no systematic review of EWS of relapse for a decade.

METHOD

We conducted a systematic review of EWS of psychotic relapse and the behavioural antecedents of worsening psychotic symptoms. Traditional EWS and ecological momentary assessment/intervention studies were included. We completed meta-analyses of the pooled sensitivity and specificity of EWS in predicting relapse, and for the prediction of relapse from individual symptoms.

RESULTS

Seventy two studies were identified including 6903 participants. Sleep, mood, and suspiciousness, emerged as predictors of worsening symptoms. Pooled sensitivity and specificity of EWS in predicting psychotic relapse was 71% and 64% (AUC value = 0.72). There was a large pooled-effect size for the model predicting relapse from individual symptom which did not reach statistical significance (d = 0.81, 95%CIs = -0.01, 1.63).

CONCLUSIONS

Important methodological advancements in the prediction of psychotic relapse in schizophrenia spectrum disorders are evident with improvements in the precision of prediction. Further efforts are required to translate these advances into effective clinical innovations.

Diagnostic performance of metagenomic next-generation sequencing for Pneumocystis jirovecii pneumonia

OBJECTIVE

Pneumocystis jirovecii pneumonia (PJP) can be a life-threatening opportunistic infection. We aimed to evaluate the diagnostic accuracy of metagenomic next-generation sequencing (mNGS) for PJP.

METHODS

A comprehensive electronic literature search of Web of Knowledge, PubMed, Cochrane Library, CNKI and Wanfang data was performed. Bivariate analysis was conducted to calculate the pooled sensitivity, specificity, diagnostic odds ratio (DOR), the area under the summary receiver operator characteristic (SROC) curve and the Q-point value (Q*).

RESULTS

The literature search resulted in 9 studies with a total of 1343 patients, including 418 cases diagnosed with PJP and 925 controls. The pooled sensitivity of mNGS for diagnosis of PJP was 0.974 [95% confidence interval (CI), 0.953-0.987]. The pooled specificity was 0.943 (95% CI, 0.926-0.957), the DOR was 431.58 (95% CI, 186.77-997.27), the area under the SROC curve was 0.987, and the Q* was 0.951. The I2 test indicated no heterogeneity between studies. The Deek funnel test suggested no potential publication bias. Subgroup analyses showed that the area under the SROC curve of mNGS for diagnosis of PJP in immunocompromised and non-HIV patients was 0.9852 and 0.979, respectively.

CONCLUSIONS

Current evidence indicates that mNGS exhibits excellent accuracy for the diagnosis of PJP. The mNGS is a promising tool for assessment of PJP in both immunocompromised and non-HIV patients.

Prediction of intraperitoneal adhesions in repeated cesarean sections: A Systematic review and Meta-analysis

OBJECTIVE

With the rate of repeated cesarean sections on the rise, intraperitoneal adhesions caused by repeated cesareans could give rise to maternal morbidity during delivery. As a result, it's critical to have the ability to predict adhesions. The current meta-analysis aims to determine if intraperitoneal adhesions are likely to be present based on the characteristics of the cesarean scar, striae gravidarum, and sliding sign.

MATERIALS AND METHODS

We systematically searched electronicdatabases beforeretrieving articles up until October 13th, 2022 for analysis. After data extraction and literature screening, we first performed a quality assessment using the QUADAS-2 score system. Following that, a bivariate random-effect meta-analysis model was used to obtain the combined diagnostic and predictive values. To pinpoint the origins of heterogeneity, we conducted a subgroup analysis. Fagan's Nomogram was used to validate the clinical utility. Sensitivity analysis was used to gauge the reliability of each included study, and Egger's test and funnel plot asymmetry was used to investigate publication bias.

RESULTS

25 studies totaling 1840 patients with intra-abdominal adhesions and 2501 controls without adhesions were included in the systematic review. Diagnostic values from 8 studies regarding skin characteristics were combined, and the results for depressed scar showed: sensitivity[95 %CI] = 0.38[0.34-0.42]; Specificity[95 %CI] = 0.88[0.85-0.90]; DOR[95 %CI] = 4.78[2.50-9.13]; AUC = 0.65. Negative sliding sign from 7 studies, although not showing a diagnostic difference between cases and controls, had excellent predictive values: sensitivity[95 %CI] = 0.71[0.65-0.77]; Specificity[95 %CI] = 0.87[0.85-0.89]; DOR[95 %CI] = 6.88[0.6-78.9]; AUC = 0.77. Subgroup analysis illustrated non-Turkish studies to reveal more significant associations than Turkish studies.

CONCLUSION

Our meta-analysis found that the occurrence of adhesions can be predicted by the characteristics of abdominal wounds, particularly depressed scar, and scar width,as well as a negative sliding sign following a previous cesarean section.

The Efficiency of Commercial Immunodiagnostic Assays for the Field Detection of Schistosoma japonicum Human Infections: A Meta-Analysis

Although great strides have been achieved, schistosomiasis japonica remains a major public health concern in China. Immunodiagnostics have been widely accepted as the first choice in large-scale screening of Schistosoma japonicum human infections, and indirect hemagglutination test (IHA), enzyme-linked immunosorbent assay (ELISA), and dipstick dye immunoassay (DDIA) are currently the three most common immunological tests for the diagnosis of S. japonicum human infections in China. This meta-analysis aimed to comprehensively assess the performance of IHA, ELISA, and DDIA for the field diagnosis of S. japonicum human infections. A total of 37 eligible publications were enrolled in the final analysis, including 29 Chinese publications and 8 English publications. No significant heterogeneities were detected among the studies reporting ELISA (I2 = 88%, p < 0.05), IHA (I2 = 95%, p < 0.05), or DDIA (I2 = 84%, p < 0.05). DDIA showed the highest pooled sensitivity (90.8%, 95% CI: 84.6% to 94.7%) and IHA presented the highest pooled specificity for detection of S. japonicum human infections (71.6%, 95% CI: 65.9% to 76.7%). Summary receiver operating characteristic (SROC) curve analysis showed that IHA exhibited the highest area under the SROC curve (AUC) (0.88, 95% CI: 0.85 to 0.9), and ELISA presented the lowest AUC (0.85, 95% CI: 0.82 to 0.88). Deeks’ funnel plots indicated no publication bias. IHA presented the highest sensitivity in medium-endemicity regions and the highest specificity for diagnosis of S. japonicum human infections in low-endemicity regions, and ELISA showed the highest diagnostic sensitivity in high-endemicity regions and the highest specificity in medium-endemicity regions, while DDIA exhibited the highest diagnostic sensitivity in high-endemicity regions and the highest specificity in low-endemicity regions. IHA and DDIA presented a higher efficiency for the diagnosis of S. japonicum human infections in marshland and lake regions than in hilly and mountainous regions, while ELISA showed a comparable diagnostic sensitivity between in marshland and lake regions and hilly and mountainous regions (88.3% vs. 88.6%), and a higher specificity in marshland and lake regions than in hilly and mountainous regions (60% vs. 48%). Our meta-analysis demonstrates a comparable diagnostic accuracy of IHA, ELISA, and DDIA for S. japonicum human infections, and the diagnostic sensitivity and specificity of IHA, ELISA, and DDIA vary in types and infection prevalence of endemic regions. DDIA combined with IHA is recommended as a tool for screening chemotherapy targets and seroepidemiological surveys during the stage moving towards schistosomiasis elimination in China. Further studies to examine the effectiveness of combinations of two or three immunological tests for diagnosis of S. japonicum human infections are warranted.

A comparison of artificial intelligence versus radiologists in the diagnosis of thyroid nodules using ultrasonography: a systematic review and meta-analysis

BACKGROUND

Thyroid nodules are common. Ultrasonography (US) is the first investigation for thyroid nodules. Artificial Intelligence (AI) is widely integrated into medical diagnosis to provide additional information. The primary objective of this study was to accumulate the pooled sensitivity and specificity between all available AI and radiologists using thyroid US imaging. The secondary objective was to compare AI's diagnostic performance to that of radiologists.

MATERIALS AND METHODS

A systematic review meta-analysis. PubMed, Scopus, Web of Science, and Cochrane Library data were searched for studies from inception until June 11, 2020.

RESULTS

Twenty five studies were included in this meta-analysis. The pooled sensitivity and specificity of AI were 0.86 (95% CI 0.81-0.91) and 0.78 (95% CI 0.73-0.83), respectively. The pooled sensitivity and specificity of radiologists were 0.85 (95% CI 0.80-0.89) and 0.82 (95% CI 0.77-0.86), respectively. The accuracy of AI and radiologists is equivalent in terms of AUC [AI 0.89 (95% CI 0.86-0.92), radiologist 0.91 (95% CI 0.88-0.93)]. The diagnostic odd ratio (DOR) between AI 23.10 (95% CI 14.20-37.58) and radiologists 27.12 (95% CI 17.45-42.16) had no statistically significant difference (P = 0.56). Meta-regression analysis revealed that Deep Learning AI had significantly greater sensitivity and specificity than classic machine learning AI (P < 0.001).

CONCLUSION

AI demonstrated comparable performance to radiologists in diagnosing benign and malignant thyroid nodules using ultrasonography. Additional research to establish its equivalency should be conducted.

Predictive validity of the Cornell Scale for depression in dementia among older adults with and without dementia: A systematic review and meta-analysis

This review analyzes the predictive validity of the Cornell Scale for Depression in Dementia (CSDD), considering cognitive function. Electronic searches were performed using MEDLINE, EMBASE, CINAHL, and PsycINFO databases. Overall, 20 studies were reviewed, including 3,499 older adults, with and without dementia, who satisfied the selection criteria. Participants were stratified into the dementia, non-dementia, and mixed groups, and the diagnostic performance of the CSDD was assessed using receiver operating characteristic (ROC) curves. In the dementia and non-dementia groups, the area under the curve was ≥ 0.9, confirming high accuracy of the test. The pooled sensitivity was the highest in the dementia group (0.87), followed by the mixed group (0.84) and non-dementia group (0.82). In the subgroup analysis based on the Mini-Mental State Examination (MMSE), the ROC curve was 0.90 for older adults with MMSE scores < 15 and 0.87 for those with an MMSE score ≥ 15. Therefore, the CSDD can be considered excellent tool for assessing depression in all older adults, although its predictive ability is better for older adults with dementia compared to that in those without dementia. It can be used as a first-line screening tool for depression, regardless of cognitive function.

Diagnostic test accuracy of pretreatment collateral score in predicting stroke outcomes after intra-arterial endovascular thrombectomy: a meta-analysis in DSA and CTA

OBJECTIVES

This study compared the diagnostic accuracy of pretreatment circulation collateral scoring (CS) system using digital subtraction angiography (DSA) and computed tomography angiography (CTA) in predicting favorable functional outcome (FFO) after intra-arterial endovascular thrombectomy (IA-EVT). Subgroup analysis characterizing scoring systems within each category was additionally conducted.

MATERIALS AND METHODS

We performed a diagnostic meta-analysis to assess the sensitivity and specificity of each CS system by using DSA and CTA, respectively. The hierarchical summary receiver operating characteristic curve (HSROC) models were used to estimate the diagnostic odds ratio (DOR) and area under the curve (AUC). The Bayes theorem was employed to determine posttest probability (PTP).

RESULTS

In total, 14 and 21 studies were assessed with DSA and CTA, respectively. In DSA, the pooled sensitivity and specificity were 0.72 (95% CI, 0.63-0.79) and 0.61 (0.53-0.68), respectively, and in the HSROC model, the DOR was 3.94 (2.71-5.73), and the AUC was 0.71 (90.67-0.75). CTA revealed a pooled sensitivity and specificity of 0.74 (0.64-0.82) and 0.53 (0.44-0.62), respectively, and in the HSROC model, the DOR was 3.17 (2.34-4.50), and the AUC was 0.67 (0.63-0.71). With a pretest probability of 26.3%, the CS in DSA and CTA exhibited limited increase of PTPs of 39% and 36%, respectively, in detecting the FFO on day 90.

CONCLUSION

DSA and CTA have comparable accuracy and are limited in predicting the functional outcome. The collateral score systems assessed with DSA and CTA were more suitable for screening than diagnosis for patients before IA-EVT.

KEY POINTS

• Our study revealed the differences of various scoring systems for assessing collateral status. • DSA and CTA have comparable accuracy, but both imaging modalities played relatively limited roles in predicting functional outcome on day 90. • The collateral score systems assessed with DSA and CTA were more suitable for screening than diagnosis for patients before IA-EVT.

Standardization of SYBR Green-Based Real-Time PCR Through the Evaluation of Different Thresholds for Different Skin Layers: An Accuracy Study and Track of the Transmission Potential of Multibacillary and Paucibacillary Leprosy Patients

The development of new molecular techniques is essential for the early diagnosis of leprosy. Studies in the field have failed to elucidate the performance of these tests in clinical practice. We aimed to design a new primer pair for the repetitive element (RLEP) target of Mycobacterium leprae and to test the accuracy of SYBR green-based real-time PCR through the evaluation of different thresholds for different skin layers. We also aimed to track the transmission potential of multibacillary and paucibacillary leprosy patients. The in vitro validation of our reaction resulted in a quantification limit of 0.03 bacilli. We then conducted a cross-sectional/cohort-based study of diagnostic accuracy. Patients were included, and skin samples were divided into four layers: epidermis, superior dermis, inferior dermis, and hypodermis. We also quantified M. leprae in nasal swabs of the included patients and compared the results to the number of household contacts also diagnosed with leprosy. One hundred patients with a clinical presentation compatible with leprosy were allocated to the leprosy or control group. Although the parasite load was greater in the superior and inferior dermis, M. leprae DNA was found in all skin layers. The best sensitivity was observed for the superior dermis using the presence of any quantifiable bacillus DNA as the threshold [sensitivity=59.26% (95% CI=45.97–71.32)]. In the epidermis, setting 1 quantifiable bacillus as the threshold resulted in 100% specificity (95% CI=92.29–100). The number of bacilli found in nasal swabs was not significantly related to the number of household contacts also diagnosed with leprosy. Paucibacillary patients tested positive only for bacillus fragments in nasal swabs but not for the entire bacilli. We can conclude that superficial biopsies might result in sensitivity loss, although different skin sample types will have little influence on the final accuracy. In contrast, threshold changes greatly influence these properties. Paucibacillary patients may not be a relevant source of disease transmission.

Paediatric tuberculosis diagnosis using Mycobacterium tuberculosis real-time polymerase chain reaction assay: a systematic review and meta-analysis

BACKGROUND

Rapid and accurate diagnosis of paediatric tuberculosis (TB) is key to manage the disease and to control and prevent its transmission. Collection of quality sputum samples without invasion methods from paediatrics (age < 16 years) with presumptive pulmonary tuberculosis (PTB) remains a challenge. Thus, the aim of this meta-analysis was to assess the overall accuracy of a real-time polymerase chain reaction (RT-PCR)-based assay, for routine diagnosis of MTB in different samples from paediatrics with active pulmonary and extra-pulmonary tuberculosis using mycobacterial culture as the gold standard in clinical microbiology laboratories.

METHODS

We conducted a systematic review and meta-analysis to examine the diagnostic test accuracy of RT-PCR based assay for the detection of MTB in paediatric clinical samples. A systematic literature search was performed for publications in any language. MEDLINE via PubMed, EMBASE, and Web of Science were among 9 bibliographic databases searched from August 2019 until November 2020. Bivariate random-effects model of meta-analysis were performed to generate pooled summary estimates (95% CIs) for overall accuracy of RT-PCR based assay compared to mycobacterial culture as the reference standard.

RESULTS

Of the 1592 candidate studies, twenty-one eligible studies met our inclusion criteria. In total, the review and meta-analysis included 5536 (3209 PTB and 2327 EPTB). Summary estimates for pulmonary TB (11 studies) were as follows: sensitivity 56 (95% CI 51-62), specificity 97 (95% CI 96-98) and summary estimates for extra-pulmonary TB (10 studies) were as follows: sensitivity 87 (95% CI 82-91)) specificity 100 (95% CI 99-100). There was significant heterogeneity in sensitivity and specificity among the enrolled studies (p < 0.001).

CONCLUSIONS

Our results suggested that the RT-PCR based assay could be a useful test for the diagnosis of paediatrics TB with high sensitivity and specificity in low-income/high-burden and upper medium income/low-burden settings. From the study, RT-PCR assay demonstrated a high degree of sensitivity for extra-pulmonary TB and good sensitivity for pulmonary TB which is an important factor in achieving effective global control and for patient management in terms of initiating early and appropriate anti-tubercular therapy.

SYSTEMATIC REVIEW REGISTRATION

PROSPERO CRD42018104052.

Ultrasonography and Fine-Needle Aspiration in Indeterminate Thyroid Nodules: A Systematic Review of Diagnostic Test Accuracy

OBJECTIVES/HYPOTHESIS

Sonographic risk criteria may assist in further prognostication of indeterminate thyroid nodules (ITNs). Our aim was to determine whether sonographic criteria could further delineate the post-test probability of malignancy in ITNs.

STUDY DESIGN

Meta-analysis of diagnostic test accuracy.

METHODS

A systematic review of Web of Science, MEDLINE, EMBASE, and CINAHL was performed from inception to April 15, 2021. Eligible studies included those which reported ultrasonographic evaluations with the American Thyroid Association (ATA) or the Thyroid Imaging Reporting and Data System (TIRADS) in adult patients with ITNs. ATA or TIRADS were scored as low (negative) or high (positive) malignancy risk using a previously validated binary classification. Primary outcomes included pooled sensitivity, specificity, likelihood ratios, and diagnostic odds ratio for all sonographic criteria. Studies were appraised using Quality Assessment of Diagnostic Accuracy Studies and the data were pooled using bivariate random-effects models.

RESULTS

Seventeen studies were included in the analysis. For Bethesda III, ATA had a specificity (0.90, 95% confidence interval (CI): 0.74-0.94), but a sensitivity of 0.52 (95% CI: 0.25-0.77). Conversely, K-TIRADS had the highest sensitivity (0.78, 95% CI: 0.62-0.89) with a specificity of 0.53 (95% CI: 0.31-0.74). Furthermore, American College of Radiology and EU TIRADS had specificities of 0.60 (95% CI: 0.36-0.80) and 0.81 (95% CI: 0.73-0.87) with sensitivities of 0.70 (95% CI: 0.37-0.90) and 0.38 (95% CI: 0.20-0.60), respectively. There were few studies with Bethesda IV nodules.

CONCLUSIONS

Though dependent on malignancy rates, Bethesda III nodules with low-suspicion TIRADS features may benefit from clinical observation, whereas nodules with high-suspicion ATA features may require molecular testing and/or surgery.

LEVEL OF EVIDENCE

NA Laryngoscope, 2021.

How useful is the center for epidemiologic studies depression scale in screening for depression in adults? An updated systematic review and meta-analysis✰

This study analyzes the performance of the Center for Epidemiologic Studies Depression Scale (CES-D) to screen for major depressive disorder (MDD) in adults. We divided adults into three groups such as community-indwelling adults, patients with chronic diseases, and psychiatric patients. Electronic searches were performed on the MEDLINE, EMBASE, CINAHL, and PsycINFO database using the following keywords: depression, depressive disorder, major, and CES-D scale. The Quality Assessment of Diagnostic Accuracy Studies-2 was applied to assess the risk of bias in diagnostic studies. We reviewed 33 studies, including 18,271 adults that met the selection criteria. In meta-analysis, the pooled sensitivity was 0.86 on community-indwelling adults, 0.85 on patients with chronic diseases and 0.85 on psychiatric patients. The pooled specificity was 0.74, 0.84, and 0.88, respectively, and the summary receiver-operating characteristic curves were 0.88, 0.91, and 0.93, respectively. The RE correlation was a negative value (-0.394) only in patients with chronic diseases, showing no heterogeneity between studies. The CES-D, which has shown high diagnostic accuracy in adults, can be recommended for use as a first-stage screener for MDD. As a result, the early application of the CES-D can lead to disease prevention in adults at risk for depression.

The Potential of Non-Invasive Biomarkers for Early Diagnosis of Asymptomatic Patients with Endometriosis

Endometriosis is a disease that affects women of reproductive age and has a significantly negative impact on their well-being. The main symptoms are dysmenorrhoea, chronic pelvic pain and infertility. In many patients the diagnostic process is very long and can take up to 8-12 years. Laparoscopy, an invasive method, is still necessary to confirm the diagnosis. Therefore, development of more effective diagnostic markers appears to be of the utmost importance for early diagnosis of endometriosis and provision of appropriate treatment. From a clinical point of view, detection of early-stage endometriosis in asymptomatic patients is an ideal situation since early diagnosis of endometriosis may delay the onset of symptoms as well as prevent progression and complications. In the meantime, Cancer Antigen 125 (CA-125) is still the most frequently studied and used marker. Other glycoproteins, growth factors and immune markers seem to play an important role. However, the search for an ideal endometriosis marker is still underway. Further studies into the pathogenesis of endometriosis will help to identify biomarkers or sets of biomarkers with the potential to improve and speed up the diagnostic process in a non-invasive way.

Diagnostic performance of serology against histologic assessment to diagnose Sjogren's syndrome: a systematic review

The objective of this review was to assess and evaluate whether the published diagnostic accuracy studies provide evidence to sustain the current diagnostic guidelines put forth by ACR/EULAR used for patients with suspected Sjögren's syndrome (SS). Literature databases, including Medline, Embase, and EBM Reviews, were searched for relevant studies on the correlation between ACR/EULAR criteria, particularly those with a direct comparison between their accuracy in diagnosing Sjögren's syndrome. We followed Cochrane, QUADAS-2, and STARD guidelines and the four-phase flow diagram by the PRISMA Statement. Reports in several languages, but only human studies were considered. Three studies assessed the accuracy of the current diagnostic tests, and these did not present adequate designs that would allow a well-supported conclusion with a high level of certainty. Due to significant clinical and methodological heterogeneity, a meta-analysis was not performed. A qualitative review of the papers was undertaken. Neither the comparative nor the non-comparative study designs permit conclusive recommendations regarding an alternative diagnostic pathway for SS. Well-designed studies of the diagnostic accuracy of SS tests are needed to validate current guidelines or to suggest changes to the current guidelines.

Review of cardiovascular imaging in the Journal of Nuclear Cardiology 2019: Positron emission tomography, computed tomography and magnetic resonance

In 2019, the Journal of Nuclear Cardiology published excellent articles pertaining to imaging in patients with cardiovascular disease. In this review we will summarize a selection of these articles to provide a concise review of the main advancements that have recently occurred in the field and provide the reader with an opportunity to review a wide selection of articles. In this first article of this 2-part series we will focus on publications dealing with positron emission tomography, computed tomography and magnetic resonance. We will specifically discuss imaging as it relates to coronary artery disease, atherosclerosis and inflammation, coronary artery calcification, cardiomyopathies, cardiac implantable electronic devices, prosthetic valves, and left ventricular assist devices. The second part of this review will place emphasis on myocardial perfusion imaging using single-photon emission computed tomography.