-
Journal of Medical Internet Research Dec 2021Interpretation of capsule endoscopy images or movies is operator-dependent and time-consuming. As a result, computer-aided diagnosis (CAD) has been applied to enhance... (Meta-Analysis)
Meta-Analysis Review
BACKGROUND
Interpretation of capsule endoscopy images or movies is operator-dependent and time-consuming. As a result, computer-aided diagnosis (CAD) has been applied to enhance the efficacy and accuracy of the review process. Two previous meta-analyses reported the diagnostic performance of CAD models for gastrointestinal ulcers or hemorrhage in capsule endoscopy. However, insufficient systematic reviews have been conducted, which cannot determine the real diagnostic validity of CAD models.
OBJECTIVE
To evaluate the diagnostic test accuracy of CAD models for gastrointestinal ulcers or hemorrhage using wireless capsule endoscopic images.
METHODS
We conducted core databases searching for studies based on CAD models for the diagnosis of ulcers or hemorrhage using capsule endoscopy and presenting data on diagnostic performance. Systematic review and diagnostic test accuracy meta-analysis were performed.
RESULTS
Overall, 39 studies were included. The pooled area under the curve, sensitivity, specificity, and diagnostic odds ratio of CAD models for the diagnosis of ulcers (or erosions) were .97 (95% confidence interval, .95-.98), .93 (.89-.95), .92 (.89-.94), and 138 (79-243), respectively. The pooled area under the curve, sensitivity, specificity, and diagnostic odds ratio of CAD models for the diagnosis of hemorrhage (or angioectasia) were .99 (.98-.99), .96 (.94-0.97), .97 (.95-.99), and 888 (343-2303), respectively. Subgroup analyses showed robust results. Meta-regression showed that published year, number of training images, and target disease (ulcers vs erosions, hemorrhage vs angioectasia) was found to be the source of heterogeneity. No publication bias was detected.
CONCLUSIONS
CAD models showed high performance for the optical diagnosis of gastrointestinal ulcer and hemorrhage in wireless capsule endoscopy.
Topics: Capsule Endoscopy; Computers; Diagnostic Tests, Routine; Hemorrhage; Humans; Ulcer
PubMed: 34904949
DOI: 10.2196/33267 -
Journal of Medical Internet Research Jul 2023Tuberculosis (TB) was the leading infectious cause of mortality globally prior to COVID-19 and chest radiography has an important role in the detection, and subsequent... (Meta-Analysis)
Meta-Analysis
BACKGROUND
Tuberculosis (TB) was the leading infectious cause of mortality globally prior to COVID-19 and chest radiography has an important role in the detection, and subsequent diagnosis, of patients with this disease. The conventional experts reading has substantial within- and between-observer variability, indicating poor reliability of human readers. Substantial efforts have been made in utilizing various artificial intelligence-based algorithms to address the limitations of human reading of chest radiographs for diagnosing TB.
OBJECTIVE
This systematic literature review (SLR) aims to assess the performance of machine learning (ML) and deep learning (DL) in the detection of TB using chest radiography (chest x-ray [CXR]).
METHODS
In conducting and reporting the SLR, we followed the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines. A total of 309 records were identified from Scopus, PubMed, and IEEE (Institute of Electrical and Electronics Engineers) databases. We independently screened, reviewed, and assessed all available records and included 47 studies that met the inclusion criteria in this SLR. We also performed the risk of bias assessment using Quality Assessment of Diagnostic Accuracy Studies version 2 (QUADAS-2) and meta-analysis of 10 included studies that provided confusion matrix results.
RESULTS
Various CXR data sets have been used in the included studies, with 2 of the most popular ones being Montgomery County (n=29) and Shenzhen (n=36) data sets. DL (n=34) was more commonly used than ML (n=7) in the included studies. Most studies used human radiologist's report as the reference standard. Support vector machine (n=5), k-nearest neighbors (n=3), and random forest (n=2) were the most popular ML approaches. Meanwhile, convolutional neural networks were the most commonly used DL techniques, with the 4 most popular applications being ResNet-50 (n=11), VGG-16 (n=8), VGG-19 (n=7), and AlexNet (n=6). Four performance metrics were popularly used, namely, accuracy (n=35), area under the curve (AUC; n=34), sensitivity (n=27), and specificity (n=23). In terms of the performance results, ML showed higher accuracy (mean ~93.71%) and sensitivity (mean ~92.55%), while on average DL models achieved better AUC (mean ~92.12%) and specificity (mean ~91.54%). Based on data from 10 studies that provided confusion matrix results, we estimated the pooled sensitivity and specificity of ML and DL methods to be 0.9857 (95% CI 0.9477-1.00) and 0.9805 (95% CI 0.9255-1.00), respectively. From the risk of bias assessment, 17 studies were regarded as having unclear risks for the reference standard aspect and 6 studies were regarded as having unclear risks for the flow and timing aspect. Only 2 included studies had built applications based on the proposed solutions.
CONCLUSIONS
Findings from this SLR confirm the high potential of both ML and DL for TB detection using CXR. Future studies need to pay a close attention on 2 aspects of risk of bias, namely, the reference standard and the flow and timing aspects.
TRIAL REGISTRATION
PROSPERO CRD42021277155; https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=277155.
Topics: Humans; Artificial Intelligence; COVID-19; Deep Learning; Radiography; Reproducibility of Results; Tuberculosis; X-Rays
PubMed: 37399055
DOI: 10.2196/43154 -
Parasite (Paris, France) 2023Serological methods should meet the needs of leishmaniasis diagnosis due to their high sensitivity and specificity, economical and adaptable rapid diagnostic test...
Serological methods should meet the needs of leishmaniasis diagnosis due to their high sensitivity and specificity, economical and adaptable rapid diagnostic test format, and ease of use. Currently, the performances of serological diagnostic tests, despite improvements with recombinant proteins, vary greatly depending on the clinical form of leishmaniasis and the endemic area. Peptide-based serological tests are promising as they could compensate for antigenic variability and improve performance, independently of Leishmania species and subspecies circulating in the endemic areas. The objective of this systematic review was to inventory all studies published from 2002 to 2022 that evaluate synthetic peptides for serological diagnosis of human leishmaniases and also to highlight the performance (e.g., sensitivity and specificity) of each peptide reported in these studies. All clinical forms of leishmaniasis, visceral and tegumentary, and all Leishmania species responsible for these diseases were considered. Following PRISMA statement recommendations, 1,405 studies were identified but only 22 articles met the selection criteria and were included in this systematic review. These original research articles described 77 different peptides, of which several have promising performance for visceral or tegumentary leishmaniasis diagnosis. This review highlights the importance of and growing interest in synthetic peptides used for serological diagnosis of leishmaniases, and their performances compared to some widely used tests with recombinant proteins.
Topics: Humans; Animals; Dogs; Leishmaniasis, Visceral; Leishmania; Serologic Tests; Leishmaniasis; Peptides; Sensitivity and Specificity; Leishmaniasis, Cutaneous; Recombinant Proteins; Antigens, Protozoan; Enzyme-Linked Immunosorbent Assay; Dog Diseases
PubMed: 37010451
DOI: 10.1051/parasite/2023011 -
Journal of Clinical Sleep Medicine :... Sep 2018Sleep disorders in most individuals remain undiagnosed and without treatment. The use of novel tools and mobile technology has the potential to increase access to... (Meta-Analysis)
Meta-Analysis
STUDY OBJECTIVES
Sleep disorders in most individuals remain undiagnosed and without treatment. The use of novel tools and mobile technology has the potential to increase access to diagnosis. The objective of this study was to perform a quantitative and qualitative analysis of the available literature evaluating the accuracy of smartphones and portable devices to screen for sleep-disordered breathing (SDB).
METHODS
A literature review was performed between February 18, 2017 and March 15, 2017. We included studies evaluating adults with SDB symptoms through the use mobile phones and/or portable devices, using standard polysomnography as a comparison. A qualitative evaluation of studies was performed with the QUADAS-2 rating. A bivariate random-effects meta-analysis was used to obtain the estimated sensitivity and specificity of screening SDB for four groups of devices: bed/mattress-based, contactless, contact with three or more sensors, and contact with fewer than three sensors. For each group, we also reported positive predictive values and negative predictive values for mild, moderate, and severe obstructive sleep apnea (OSA) screening.
RESULTS
Of the 22 included studies, 18 were pooled in the meta-analysis. Devices that were bed/mattress-based were found to have the best sensitivity overall (0.921, 95% confidence interval [CI] 0.870, 0.953). The sensitivity of contactless devices to detect mild OSA cases was the highest of all groups (0.976, 95% CI 0.899, 0.995), but provided a high false positive rate (0.487, 95% CI 0.137, 0.851). The remaining groups of devices showed low sensitivity and heterogeneous results.
CONCLUSIONS
This study evidenced the limitations and potential use of portable devices in screening patients for SDB. Additional research should evaluate the accuracy of devices when used at home.
Topics: Humans; Monitoring, Ambulatory; Polysomnography; Sensitivity and Specificity; Sleep Apnea Syndromes; Smartphone
PubMed: 30176971
DOI: 10.5664/jcsm.7346 -
Computational Intelligence and... 2022Epileptic seizure is one of the most chronic neurological diseases that instantaneously disrupts the lifestyle of affected individuals. Toward developing novel and... (Review)
Review
Epileptic seizure is one of the most chronic neurological diseases that instantaneously disrupts the lifestyle of affected individuals. Toward developing novel and efficient technology for epileptic seizure management, recent diagnostic approaches have focused on developing machine/deep learning model (ML/DL)-based electroencephalogram (EEG) methods. Importantly, EEG's noninvasiveness and ability to offer repeated patterns of epileptic-related electrophysiological information have motivated the development of varied ML/DL algorithms for epileptic seizure diagnosis in the recent years. However, EEG's low amplitude and nonstationary characteristics make it difficult for existing ML/DL models to achieve a consistent and satisfactory diagnosis outcome, especially in clinical settings, where environmental factors could hardly be avoided. Though several recent works have explored the use of EEG-based ML/DL methods and statistical feature for seizure diagnosis, it is unclear what the advantages and limitations of these works are, which might preclude the advancement of research and development in the field of epileptic seizure diagnosis and appropriate criteria for selecting ML/DL models and statistical feature extraction methods for EEG-based epileptic seizure diagnosis. Therefore, this paper attempts to bridge this research gap by conducting an extensive systematic review on the recent developments of EEG-based ML/DL technologies for epileptic seizure diagnosis. In the review, current development in seizure diagnosis, various statistical feature extraction methods, ML/DL models, their performances, limitations, and core challenges as applied in EEG-based epileptic seizure diagnosis were meticulously reviewed and compared. In addition, proper criteria for selecting appropriate and efficient feature extraction techniques and ML/DL models for epileptic seizure diagnosis were also discussed. Findings from this study will aid researchers in deciding the most efficient ML/DL models with optimal feature extraction methods to improve the performance of EEG-based epileptic seizure detection.
Topics: Algorithms; Deep Learning; Electroencephalography; Epilepsy; Humans; Seizures; Signal Processing, Computer-Assisted; Support Vector Machine
PubMed: 35755757
DOI: 10.1155/2022/6486570 -
Thyroid : Official Journal of the... May 2015Thyroid nodules are a common finding in the general population, and their detection is increasing with the widespread use of ultrasound (US). Thyroid cancer is found in... (Meta-Analysis)
Meta-Analysis Review
BACKGROUND
Thyroid nodules are a common finding in the general population, and their detection is increasing with the widespread use of ultrasound (US). Thyroid cancer is found in 5-15% of cases depending on sex, age, and exposure to other risk factors. Some US parameters have been associated with increased risk of malignancy. However, no characteristic seems sufficiently reliable in isolation to diagnose malignancy. The objective of this meta-analysis was to evaluate the diagnostic performance of US features for thyroid malignancy in patients with unselected thyroid nodules and nodules with indeterminate fine-needle aspiration (FNA) cytology.
METHODS
Electronic databases were reviewed for studies published prior to July 2012 that evaluated US features of thyroid nodules and reported postoperative histopathologic diagnosis. A manual search of references of review and key articles, and previous meta-analyses was also performed. A separate meta-analysis was performed including only nodules with indeterminate cytology. Analyzed features were solid structure, hypoechogenicity, irregular margins, absence of halo, microcalcifications, central vascularization, solitary nodule, heterogeneity, taller than wide shape, and absence of elasticity.
RESULTS
Fifty-two observational studies (12,786 nodules) were included. Nine studies included nodules with indeterminate cytology as a separate category, comprising 1851 nodules. In unselected nodules, all US features were significantly associated with malignancy with an odds ratio varying from 1.78 to 35.7, and microcalcifications, irregular margins, and a taller than wide shape had high specificities (Sp; 87.8%, 83.1%, 96.6%) and positive likelihood ratios (LHR; 3.26, 2.99, 8.07). Absence of elasticity was the single feature with the best diagnostic performance (sensitivity 87.9%, Sp 86.2%, and positive LHR 6.39). The presence of central vascularization was the most specific US feature in nodules with indeterminate cytology (Sp 96% and positive LHR 2.13).
CONCLUSIONS
US features in isolation do not provide reliable information to select nodules that should have a FNA performed. A combination of US characteristics with higher likelihood ratios and consequently with higher post-test probabilities of malignancy-microcalcifications, or a taller than wide shape, or irregular margins, or absence of elasticity-will probably identify nodules with an increased risk for malignancy. Further studies are required to standardize elastography techniques and evaluate outcomes, especially in nodules with an indeterminate cytology.
Topics: Carcinoma; Diagnosis, Differential; Humans; Risk; Thyroid Gland; Thyroid Neoplasms; Thyroid Nodule; Ultrasonography
PubMed: 25747526
DOI: 10.1089/thy.2014.0353 -
Journal of Ultrasound Jun 2023The goal of this study was to perform a comprehensive meta-analysis to assess the overall diagnostic value of Doppler twinkling for the diagnosis of urolithiasis. (Meta-Analysis)
Meta-Analysis
OBJECTIVE
The goal of this study was to perform a comprehensive meta-analysis to assess the overall diagnostic value of Doppler twinkling for the diagnosis of urolithiasis.
METHODS
We systematically searched the PubMed, EMBASE, and Cochrane Library databases from inception through May 31, 2021. Studies including patients with urolithiasis who underwent color flow Doppler sampling to highlight the twinkling artifact and computed tomography were included. Diagnostic test meta-analysis was performed with a bivariate model. We used summary receiver operating characteristic curves to summarize the overall diagnostic performance. The weighted sensitivity, specificity, positive likelihood ratio, and negative likelihood ratio were calculated.
RESULTS
Sixteen studies involving 4572 patients were included in the systematic review and meta-analysis. The weighted sensitivity was 0.86 (95% confidence interval [CI] 0.72-0.94), specificity 0.92 (95% CI 0.75-0.98), positive likelihood ratio 11.3, negative likelihood ratio 0.2, and diagnostic odds ratio 75.5.
CONCLUSION
The Doppler twinkling artifact has good diagnostic value for the diagnosis of urolithiasis and should be used as a complementary tool in the diagnosis of urolithiasis.
Topics: Humans; Artifacts; Sensitivity and Specificity; Urolithiasis; Ultrasonography, Doppler; ROC Curve
PubMed: 36705851
DOI: 10.1007/s40477-022-00759-z -
International Journal of Clinical... 2022Distinguishing between benign and malignant thyroid nodules remains difficult. Ultrasound has been established as a non-invasive and relatively simple imaging technique... (Meta-Analysis)
Meta-Analysis
BACKGROUND
Distinguishing between benign and malignant thyroid nodules remains difficult. Ultrasound has been established as a non-invasive and relatively simple imaging technique for thyroid nodules. This study aimed to assess the diagnostic accuracy of conventional ultrasound and ultrasound elastography for the differentiation between benign and malignant thyroid nodules by meta-analyzing published studies.
METHODS
Literature was retrieved from the PubMed and Embase databases from inception to May 31, 2022. The literature was screened using inclusion and exclusion criteria. The Quality Assessment of Diagnostic Accuracy Studies (QUADAS2) scale was used to assess the quality of the included literature. Publication bias of the included studies was assessed by Deek's funnel plot. Heterogeneity tests were performed using Cochrane statistic and I statistic.
RESULTS
Finally, 9 articles were included. The meta-analysis showed that the combined sensitivity and specificity of ultrasound for the diagnosis of thyroid nodules were 0.88 [95% CI (0.83-0.91)] and 0.86 [95% CI (0.79-0.90)], respectively. The area under the curve (AUC) of the summary receiver operating characteristic curve (SROC) was 0.92 [95% CI (0.90-0.94)]. There was no significant publication bias in this study. . Existing evidence shows that ultrasound has a certain accuracy in diagnosing benign and malignant thyroid nodules, providing a scientific basis for thyroid assessment and diagnosis.
Topics: Diagnosis, Differential; Elasticity Imaging Techniques; Humans; ROC Curve; Sensitivity and Specificity; Thyroid Nodule
PubMed: 36160289
DOI: 10.1155/2022/5056082 -
Orthopaedic Journal of Sports Medicine May 2017Femoroacetabular impingement (FAI) is a well-recognized condition that causes hip pain and can lead to early osteoarthritis if not managed properly. With the increasing... (Review)
Review
BACKGROUND
Femoroacetabular impingement (FAI) is a well-recognized condition that causes hip pain and can lead to early osteoarthritis if not managed properly. With the increasing awareness and efficacy of operative treatments for pincer-type FAI, there is a need for consensus on the standardized radiographic diagnosis.
PURPOSE
To perform a systematic review of the evidence regarding imaging modalities and radiographic signs for diagnosing pincer-type FAI.
STUDY DESIGN
Systematic review; Level of evidence, 4.
METHODS
A literature review was performed in 2016 using the Cochrane, PubMed, and Embase search engines. All articles focusing on a radiographic diagnosis of pincer-type FAI were reviewed. Each of the included 44 articles was assigned the appropriate level of evidence, and the particular radiographic marker and/or type of imaging were also summarized.
RESULTS
There were 44 studies included in the final review. Most of the articles were level 4 evidence (26 articles), and there were 12 level 3 and 6 level 2 articles. The crossover sign was the most commonly used radiographic sign (27/44) followed by the lateral center-edge angle (22/44). Anteroposterior (AP) pelvis plain radiographs were the most commonly used imaging modality (33 studies). Poor-quality evidence exists in support of most currently used radiographic markers, including the crossover sign, lateral center-edge angle, posterior wall sign, ischial spine sign, coxa profunda, acetabular protrusion, and acetabular index. There is poor-quality conflicting evidence regarding the use of the herniation pit to diagnose pincer-type FAI. Some novel measurements, such as β-angle, acetabular roof ratio, and acetabular retroversion index, have been proposed, but they also lack support from the literature.
CONCLUSION
No strong evidence exists to support a single best set of current radiographic markers for the diagnosis of pincer-type FAI, largely due to the lack of better quality trials (levels 1 and 2) that compare conventional radiographic findings with the gold standard, which is the intraoperative findings. More sophisticated imaging modalities such as computed tomography and magnetic resonance arthrography are often needed to diagnose pincer-type FAI, and these investigations are relatively accurate in assessing labral pathology or cartilage damage.
PubMed: 28607941
DOI: 10.1177/2325967117708307 -
Eye (London, England) Jul 2023Cerebral Visual Impairment (CVI) is a common condition in the UK. Patients with conditions associated with CVI are frequently seen in paediatric ophthalmology clinics... (Review)
Review
Cerebral Visual Impairment (CVI) is a common condition in the UK. Patients with conditions associated with CVI are frequently seen in paediatric ophthalmology clinics offering eye care professionals an opportunity to identify children proactively. In most cases CVI occurs as part of a neurodevelopmental condition or as a feature of multiple and complex disabilities. However, CVI can also be seen in children with apparently typical development. In some cases, high contrast visual acuity is normal and in other cases severely impaired. As such, identification of CVI requires evaluation of aspects of visual performance beyond high contrast acuity and consideration that visual function of those with CVI may fluctuate. Few paediatric ophthalmologists have received formal training in CVI. The detection and diagnosis of CVI varies across the UK and patients report hugely different experiences. A diagnosis of CVI is made based on professional clinical judgement and it is recognised that individual perspectives and local practice in the specific methodologies of assessment will vary. A systematic review and survey of professionals is underway to attempt to reach agreement on diagnostic criteria. Nonetheless, established pathways and published protocols can offer guidance on how a paediatric ophthalmology service can approach assessment of the child with suspected CVI. The purpose of this paper is to present a summary of research and clinical practice methods for detecting and diagnosing CVI in a paediatric ophthalmology outpatient setting. It represents current understanding of the topic and acknowledges the evolving nature of both practice and the evidence-base. A rapid literature review was undertaken to identify articles relating to clinical investigation of children with CVI. A focus group of QTVI and subject matter experts from sight loss charities was undertaken to address areas which were not covered by the literature review.
Topics: Child; Humans; Consensus; Vision Disorders; Visual Acuity; Ophthalmology; Blindness
PubMed: 36258009
DOI: 10.1038/s41433-022-02261-6