-
JAMA Network Open Mar 2023Artificial intelligence (AI) enables powerful models for establishment of clinical diagnostic and prognostic tools for hip fractures; however the performance and... (Meta-Analysis)
Meta-Analysis
IMPORTANCE
Artificial intelligence (AI) enables powerful models for establishment of clinical diagnostic and prognostic tools for hip fractures; however the performance and potential impact of these newly developed algorithms are currently unknown.
OBJECTIVE
To evaluate the performance of AI algorithms designed to diagnose hip fractures on radiographs and predict postoperative clinical outcomes following hip fracture surgery relative to current practices.
DATA SOURCES
A systematic review of the literature was performed using the MEDLINE, Embase, and Cochrane Library databases for all articles published from database inception to January 23, 2023. A manual reference search of included articles was also undertaken to identify any additional relevant articles.
STUDY SELECTION
Studies developing machine learning (ML) models for the diagnosis of hip fractures from hip or pelvic radiographs or to predict any postoperative patient outcome following hip fracture surgery were included.
DATA EXTRACTION AND SYNTHESIS
This study followed the Preferred Reporting Items for Systematic Reviews and Meta-analyses and was registered with PROSPERO. Eligible full-text articles were evaluated and relevant data extracted independently using a template data extraction form. For studies that predicted postoperative outcomes, the performance of traditional predictive statistical models, either multivariable logistic or linear regression, was recorded and compared with the performance of the best ML model on the same out-of-sample data set.
MAIN OUTCOMES AND MEASURES
Diagnostic accuracy of AI models was compared with the diagnostic accuracy of expert clinicians using odds ratios (ORs) with 95% CIs. Areas under the curve for postoperative outcome prediction between traditional statistical models (multivariable linear or logistic regression) and ML models were compared.
RESULTS
Of 39 studies that met all criteria and were included in this analysis, 18 (46.2%) used AI models to diagnose hip fractures on plain radiographs and 21 (53.8%) used AI models to predict patient outcomes following hip fracture surgery. A total of 39 598 plain radiographs and 714 939 hip fractures were used for training, validating, and testing ML models specific to diagnosis and postoperative outcome prediction, respectively. Mortality and length of hospital stay were the most predicted outcomes. On pooled data analysis, compared with clinicians, the OR for diagnostic error of ML models was 0.79 (95% CI, 0.48-1.31; P = .36; I2 = 60%) for hip fracture radiographs. For the ML models, the mean (SD) sensitivity was 89.3% (8.5%), specificity was 87.5% (9.9%), and F1 score was 0.90 (0.06). The mean area under the curve for mortality prediction was 0.84 with ML models compared with 0.79 for alternative controls (P = .09).
CONCLUSIONS AND RELEVANCE
The findings of this systematic review and meta-analysis suggest that the potential applications of AI to aid with diagnosis from hip radiographs are promising. The performance of AI in diagnosing hip fractures was comparable with that of expert radiologists and surgeons. However, current implementations of AI for outcome prediction do not seem to provide substantial benefit over traditional multivariable predictive statistics.
Topics: Humans; Artificial Intelligence; Hip Fractures; Prognosis; Algorithms; Length of Stay
PubMed: 36930153
DOI: 10.1001/jamanetworkopen.2023.3391 -
Journal of the American Society of... Jul 2017Pulmonary embolism (PE) is a common diagnosis with significant mortality if not appropriately treated. The use of transthoracic echocardiography in patients with PE is... (Meta-Analysis)
Meta-Analysis Review
OBJECTIVE
Pulmonary embolism (PE) is a common diagnosis with significant mortality if not appropriately treated. The use of transthoracic echocardiography in patients with PE is common; however, its diagnostic capabilities in this use are unclear. With the increased use of ultrasonography in medical settings, it is important to understand the strengths and limitations of echocardiography for the diagnosis of PE.
METHODS
We conducted a systematic review of PubMed, CINAHL, and EMBASE through 2016 for articles assessing the diagnostic accuracy of transthoracic echocardiography for PE. Two authors independently abstracted relevant data from the studies. We assessed quality using the QUADAS-2 tool for diagnostic studies.
RESULTS
Undefined "right heart strain" was the most common sign used, and it had a sensitivity of 53% (95% CI, 45%-61%) and a specificity of 83% (95% CI, 74%-90%). Eleven other distinct signs were identified: ventricle size ratio, abnormal septal motion, tricuspid regurgitation, 60/60 sign, McConnell's sign, right heart thrombus, right ventricle hypokinesis, pulmonary hypertension, right ventricular end-diastolic diameter, tricuspid annular plane systolic excursion, and right ventricular systolic pressure.
CONCLUSIONS
Studies show a consistently high specificity and low sensitivity for echocardiography in the diagnosis of PE, making it potentially adequate as a rule-in test at the bedside in critical care settings such as the emergency department and intensive care unit for patients with a suspicion of PE, especially those unable to get other confirmatory studies. Future research may continue to clarify the role of bedside echocardiography in conjunction with other tests and imaging in the overall management of PE.
Topics: Diagnosis, Differential; Echocardiography; Humans; Pulmonary Embolism
PubMed: 28495379
DOI: 10.1016/j.echo.2017.03.004 -
BMJ Open Nov 2019To estimate the prevalence and incidence of placenta previa complicated by placenta accreta spectrum (PAS) and to examine the different criteria being used for the... (Meta-Analysis)
Meta-Analysis
OBJECTIVE
To estimate the prevalence and incidence of placenta previa complicated by placenta accreta spectrum (PAS) and to examine the different criteria being used for the diagnosis.
DESIGN
Systematic review and meta-analysis.
DATA SOURCES
PubMed, Google Scholar, ClinicalTrials.gov and MEDLINE were searched between August 1982 and September 2018.
ELIGIBILITY CRITERIA
Studies reporting on placenta previa complicated by PAS diagnosed in a defined obstetric population.
DATA EXTRACTION AND SYNTHESIS
Two independent reviewers performed the data extraction using a predefined protocol and assessed the risk of bias using the Newcastle-Ottawa scale for observational studies, with difference agreed by consensus. The primary outcomes were overall prevalence of placenta previa, incidence of PAS according to the type of placenta previa and the reported clinical outcomes, including the number of peripartum hysterectomies and direct maternal mortality. The secondary outcomes included the criteria used for the prenatal ultrasound diagnosis of placenta previa and the criteria used to diagnose and grade PAS at birth.
RESULTS
A total of 258 articles were reviewed and 13 retrospective and 7 prospective studies were included in the analysis, which reported on 587 women with placenta previa and PAS. The meta-analysis indicated a significant (p<0.001) heterogeneity between study estimates for the prevalence of placenta previa, the prevalence of placenta previa with PAS and the incidence of PAS in the placenta previa cohort. The median prevalence of placenta previa was 0.56% (IQR 0.39-1.24) whereas the median prevalence of placenta previa with PAS was 0.07% (IQR 0.05-0.16). The incidence of PAS in women with a placenta previa was 11.10% (IQR 7.65-17.35).
CONCLUSIONS
The high heterogeneity in qualitative and diagnostic data between studies emphasises the need to implement standardised protocols for the diagnoses of both placenta previa and PAS, including the type of placenta previa and grade of villous invasiveness.
PROSPERO REGISTRATION NUMBER
CRD42017068589.
Topics: Female; Humans; Hysterectomy; Incidence; Peripartum Period; Placenta Accreta; Placenta Previa; Pregnancy; Prevalence; Ultrasonography, Prenatal
PubMed: 31722942
DOI: 10.1136/bmjopen-2019-031193 -
The Journal of Prosthetic Dentistry Dec 2023Artificial intelligence (AI) models have been developed for periodontal applications, including diagnosing gingivitis and periodontal disease, but their accuracy and... (Review)
Review
STATEMENT OF PROBLEM
Artificial intelligence (AI) models have been developed for periodontal applications, including diagnosing gingivitis and periodontal disease, but their accuracy and maturity of the technology remain unclear.
PURPOSE
The purpose of this systematic review was to evaluate the performance of the AI models for detecting dental plaque and diagnosing gingivitis and periodontal disease.
MATERIAL AND METHODS
A review was performed in 4 databases: MEDLINE/PubMed, World of Science, Cochrane, and Scopus. A manual search was also conducted. Studies were classified into 4 groups: detecting dental plaque, diagnosis of gingivitis, diagnosis of periodontal disease from intraoral images, and diagnosis of alveolar bone loss from periapical, bitewing, and panoramic radiographs. Two investigators evaluated the studies independently by applying the Joanna Briggs Institute critical appraisal. A third examiner was consulted to resolve any lack of consensus.
RESULTS
Twenty-four articles were included: 2 studies developed AI models for detecting plaque, resulting in accuracy ranging from 73.6% to 99%; 7 studies assessed the ability to diagnose gingivitis from intraoral photographs reporting an accuracy between 74% and 78.20%; 1 study used fluorescent intraoral images to diagnose gingivitis reporting 67.7% to 73.72% accuracy; 3 studies assessed the ability to diagnose periodontal disease from intraoral photographs with an accuracy between 47% and 81%, and 11 studies evaluated the performance of AI models for detecting alveolar bone loss from radiographic images reporting an accuracy between 73.4% and 99%.
CONCLUSIONS
AI models for periodontology applications are still in development but might provide a powerful diagnostic tool.
Topics: Humans; Dental Plaque; Alveolar Bone Loss; Artificial Intelligence; Periodontal Diseases; Gingivitis
PubMed: 35300850
DOI: 10.1016/j.prosdent.2022.01.026 -
Critical Care (London, England) Sep 2014The understanding of coagulopathies in trauma has increased interest in thromboelastography (TEG®) and thromboelastometry (ROTEM®), which promptly evaluate the entire... (Review)
Review
Effect of thromboelastography (TEG®) and rotational thromboelastometry (ROTEM®) on diagnosis of coagulopathy, transfusion guidance and mortality in trauma: descriptive systematic review.
INTRODUCTION
The understanding of coagulopathies in trauma has increased interest in thromboelastography (TEG®) and thromboelastometry (ROTEM®), which promptly evaluate the entire clotting process and may guide blood product therapy. Our objective was to review the evidence for their role in diagnosing early coagulopathies, guiding blood transfusion, and reducing mortality in injured patients.
METHODS
We considered observational studies and randomized controlled trials (MEDLINE, EMBASE, and Cochrane databases) to February 2014 that examined TEG®/ROTEM® in adult trauma patients. We extracted data on demographics, diagnosis of early coagulopathies, blood transfusion, and mortality. We assessed methodologic quality by using the Newcastle-Ottawa scale (NOS) for observational studies and QUADAS-2 tool for diagnostic accuracy studies.
RESULTS
Fifty-five studies (12,489 patients) met inclusion criteria, including 38 prospective cohort studies, 15 retrospective cohort studies, two before-after studies, and no randomized trials. Methodologic quality was moderate (mean NOS score, 6.07; standard deviation, 0.49). With QUADAS-2, only three of 47 studies (6.4%) had a low risk of bias in all domains (patient selection, index test, reference standard and flow and timing); 37 of 47 studies (78.8%) had low concerns regarding applicability. Studies investigated TEG®/ROTEM® for diagnosis of early coagulopathies (n = 40) or for associations with blood-product transfusion (n = 25) or mortality (n = 24). Most (n = 52) were single-center studies. Techniques examined included rapid TEG® (n =12), ROTEM® (n = 18), TEG® (n = 23), or both TEG® and rapid TEG® (n = 2). Many TEG®/ROTEM® measurements were associated with early coagulopathies, including some (hypercoagulability, hyperfibrinolysis, platelet dysfunction) not assessed by routine screening coagulation tests. Standard measures of diagnostic accuracy were inconsistently reported. Many abnormalities predicted the need for massive transfusion and death, but predictive performance was not consistently superior to routine tests. One observational study suggested that a ROTEM®-based transfusion algorithm reduced blood-product transfusion, but TEG®/ROTEM®-based resuscitation was not associated with lower mortality in most studies.
CONCLUSIONS
Limited evidence from observational data suggest that TEG®/ROTEM® tests diagnose early trauma coagulopathy and may predict blood-product transfusion and mortality in trauma. Effects on blood-product transfusion, mortality, and other patient-important outcomes remain unproven in randomized trials.
Topics: Blood Coagulation Disorders; Blood Transfusion; Humans; Thrombelastography; Wounds and Injuries
PubMed: 25261079
DOI: 10.1186/s13054-014-0518-9 -
Acta Orthopaedica Et Traumatologica... 2016The purpose of this study was to evaluate the sensitivity and specificity of 3 tests for assessing anterior cruciate ligament (ACL) ruptures. (Meta-Analysis)
Meta-Analysis
OBJECTIVE
The purpose of this study was to evaluate the sensitivity and specificity of 3 tests for assessing anterior cruciate ligament (ACL) ruptures.
METHODS
MEDLINE, EMBASE, Cochrane Library, and CBM (Chinese Biomedical Literature Database) searches were performed. Studies selected for data extraction were those that addressed the accuracy of at least 1 physical diagnostic test for ACL rupture in comparison with a clinical reference standard such as arthroscopy, arthrotomy, or magnetic resonance imaging (MRI). The references of the included studies were also reviewed. Searches were limited to English and Chinese languages.
RESULTS
Sixteen studies that assessed the accuracy of the 3 tests for diagnosing ACL ruptures met the inclusion criteria. Study results were, however, heterogeneous. The Lachman test is the most sensitive test to determine ACL tears, showing a pooled sensitivity of 87.1% (95% confidence interval [CI] 0.84-0.90). The pivot shift test is the most specific test, showing a pooled specificity of 97.5% (95% CI 0.95-0.99); additionally, it has the highest positive likelihood ratios (LR+) of 16.00 (95% CI 7.34-34.87). The Lachman test has the lowest negative likelihood ratios (LR-) of 0.17 (95% CI 0.11-0.25).
CONCLUSION
In cases of suspected ACL injury, it is recommended to perform the pivot shift test, as it is highly specific and has greater likelihood and discrimination of accurately diagnosing ACL rupture. The Lachman test has great efficacy in ruling out a diagnosis of ACL rupture because of the lowest negative likelihood ratios.
Topics: Anterior Cruciate Ligament; Anterior Cruciate Ligament Injuries; Arthroscopy; Comparative Effectiveness Research; Dimensional Measurement Accuracy; Humans; Magnetic Resonance Imaging; Sensitivity and Specificity
PubMed: 26854045
DOI: 10.3944/AOTT.2016.14.0283 -
Chiropractic & Manual Therapies Dec 2022Clinicians rely on certain physical examination tests to diagnose and potentially grade ankle sprains and ankle instability. Diagnostic error and inaccurate prognosis...
INTRODUCTION
Clinicians rely on certain physical examination tests to diagnose and potentially grade ankle sprains and ankle instability. Diagnostic error and inaccurate prognosis may have important repercussions for clinical decision-making and patient outcomes. Therefore, it is important to recognize the diagnostic value of orthopaedic tests through understanding the reliability and validity of these tests.
OBJECTIVE
To systematically review and report evidence on the reliability and validity of orthopaedic tests for the diagnosis of ankle sprains and instability.
METHODS
PubMed, CINAHL, Scopus, and Cochrane databases were searched from inception to December 2021. In addition, the reference list of included studies, located systematic reviews, and orthopaedic textbooks were searched. All articles reporting reliability or validity of physical examination or orthopaedic tests to diagnose ankle instability or sprains were included. Methodological quality of the reliability and the validity studies was assessed with The Quality Appraisal for Reliability studies checklist and the Quality Assessment of Diagnostic Accuracy Studies-2 respectively. We identified the number of times the orthopaedic test was investigated and the validity and/or reliability of each test.
RESULTS
Overall, sixteen studies were included. Three studies assessed reliability, eight assessed validity, and five evaluated both. Overall, fifteen tests were evaluated, none demonstrated robust reliability and validity scores. The anterolateral talar palpation test reported the highest diagnostic accuracy. Further, the anterior drawer test, the anterolateral talar palpation, the reverse anterior lateral drawer test, and palpation of the anterior talofibular ligament reported the highest sensitivity. The highest specificity was attributed to the anterior drawer test, the anterolateral drawer test, the reverse anterior lateral drawer test, tenderness on palpation of the proximal fibular, and the squeeze test.
CONCLUSION
Overall, the diagnostic accuracy, reliability, and validity of physical examination tests for the assessment of ankle instability were limited. Physical examination tests should not be used in isolation, but rather in combination with the clinical history to diagnose an ankle sprain. Preliminary evidence suggests that the overall validity of physical examination for the ankle may be better if conducted five days after the injury rather than within 48 h of injury.
Topics: Humans; Ankle; Reproducibility of Results; Ankle Joint; Physical Examination; Joint Instability; Ankle Injuries
PubMed: 36536446
DOI: 10.1186/s12998-022-00470-0 -
Journal of Foot and Ankle Research 2018Flexible flat foot is a normal observation in typically developing children, however, some children with flat feet present with pain and impaired lower limb function.... (Review)
Review
BACKGROUND
Flexible flat foot is a normal observation in typically developing children, however, some children with flat feet present with pain and impaired lower limb function. The challenge for health professionals is to identify when foot posture is outside of expected findings and may warrant intervention. Diagnoses of flexible flat foot is often based on radiographic or clinical measures, yet the validity and reliability of these measures for a paediatric population is not clearly understood. The aim of this systematic review was to investigate how paediatric foot posture is defined and measured within the literature, and if the psychometric properties of these measures support any given diagnoses.
METHODS
Electronic databases (MEDLINE, CINAHL, EMBASE, Cochrane, AMED, SportDiscus, PsycINFO, and Web of Science) were systematically searched in January 2017 for empirical studies where participants had diagnosed flexible flat foot and were aged 18 years or younger. Outcomes of interest were the foot posture measures and definitions used. Further articles were sought where cited in relation to the psychometric properties of the measures used.
RESULTS
Of the 1101 unique records identified by the searches, 27 studies met the inclusion criteria involving 20 foot posture measures and 40 definitions of paediatric flexible flat foot. A further 18 citations were sought in relation to the psychometric properties of these measures. Three measures were deemed valid and reliable, the FPI-6 > + 6 for children aged three to 15 years, a Staheli arch index of > 1.07 for children aged three to six and ≥ 1.28 for children six to nine, and a Chippaux-Smirak index of > 62.7% in three to seven year olds, > 59% in six to nine year olds and ≥ 40% for children aged nine to 16 years. No further measures were found to be valid for the paediatric population.
CONCLUSION
No universally accepted criteria for diagnosing paediatric flat foot was found within existing literature, and psychometric data for foot posture measures and definitions used was limited. The outcomes of this review indicate that the FPI - 6, Staheli arch index or Chippaux-Smirak index should be the preferred method of paediatric foot posture measurement in future research.
Topics: Anthropometry; Child; Child Development; Flatfoot; Foot; Humans; Posture; Psychometrics; Reproducibility of Results; Research Design
PubMed: 29854006
DOI: 10.1186/s13047-018-0264-3 -
The Spine Journal : Official Journal of... Jan 2018In clinical practice, the diagnosis of cervical radiculopathy is based on information from the patient's history, physical examination, and diagnostic imaging. Various... (Meta-Analysis)
Meta-Analysis Review
BACKGROUND CONTEXT
In clinical practice, the diagnosis of cervical radiculopathy is based on information from the patient's history, physical examination, and diagnostic imaging. Various physical tests may be performed, but their diagnostic accuracy is unknown.
PURPOSE
This study aimed to summarize and update the evidence on diagnostic performance of tests carried out during a physical examination for the diagnosis of cervical radiculopathy.
STUDY DESIGN
A review of the accuracy of diagnostic tests was carried out.
STUDY SAMPLE
The study sample comprised diagnostic studies comparing results of tests performed during a physical examination in diagnosing cervical radiculopathy with a reference standard of imaging or surgical findings.
OUTCOME MEASURES
Sensitivity, specificity, likelihood ratios are presented, together with pooled results for sensitivity and specificity.
METHODS
A literature search up to March 2016 was performed in CENTRAL, PubMed (MEDLINE), Embase, CINAHL, Web of Science, and Google Scholar. The methodological quality of studies was assessed using the QUADAS-2.
RESULTS
Five diagnostic accuracy studies were identified. Only Spurling's test was evaluated in more than one study, showing high specificity ranging from 0.89 to 1.00 (95% confidence interval [CI]: 0.59-1.00); sensitivity varied from 0.38 to 0.97 (95% CI: 0.21-0.99). No studies were found that assessed the diagnostic accuracy of widely used neurological tests such as key muscle strength, tendon reflexes, and sensory impairments.
CONCLUSIONS
There is limited evidence for accuracy of physical examination tests for the diagnosis of cervical radiculopathy. When consistent with patient history, clinicians may use a combination of Spurling's, axial traction, and an Arm Squeeze test to increase the likelihood of a cervical radiculopathy, whereas a combined results of four negative neurodynamics tests and an Arm Squeeze test could be used to rule out the disorder.
Topics: Cervical Vertebrae; Humans; Neurologic Examination; Radiculopathy; Sensitivity and Specificity
PubMed: 28838857
DOI: 10.1016/j.spinee.2017.08.241 -
JAMA Network Open Mar 2019Standard tools used to diagnose pulmonary edema in acute decompensated heart failure (ADHF), including chest radiography (CXR), lack adequate sensitivity, which may... (Meta-Analysis)
Meta-Analysis
Diagnostic Accuracy of Point-of-Care Lung Ultrasonography and Chest Radiography in Adults With Symptoms Suggestive of Acute Decompensated Heart Failure: A Systematic Review and Meta-analysis.
IMPORTANCE
Standard tools used to diagnose pulmonary edema in acute decompensated heart failure (ADHF), including chest radiography (CXR), lack adequate sensitivity, which may delay appropriate diagnosis and treatment. Point-of-care lung ultrasonography (LUS) may be more accurate than CXR, but no meta-analysis of studies directly comparing the 2 tools was previously available.
OBJECTIVE
To compare the accuracy of LUS with the accuracy of CXR in the diagnosis of cardiogenic pulmonary edema in adult patients presenting with dyspnea.
DATA SOURCES
A comprehensive search of MEDLINE, Embase, and Cochrane Library databases and the gray literature was performed in May 2018. No language or year limits were applied.
STUDY SELECTION
Study inclusion criteria were a prospective adult cohort of patients presenting to any clinical setting with dyspnea who underwent both LUS and CXR on initial assessment with imaging results compared with a reference standard ADHF diagnosis by a clinical expert after either a medical record review or a combination of echocardiography findings and brain-type natriuretic peptide criteria. Two reviewers independently assessed the studies for inclusion criteria, and disagreements were resolved with discussion.
DATA EXTRACTION AND SYNTHESIS
Reporting adhered to the Cochrane Handbook for Systematic Reviews of Diagnostic Test Accuracy and the Preferred Reporting Items for Systematic Reviews and Meta-analyses guidelines. Two authors independently extracted data and assessed the risk of bias using a customized QUADAS-2 tool. The pooled sensitivity and specificity of LUS and CXR were determined using a hierarchical summary receiver operating characteristic approach.
MAIN OUTCOMES AND MEASURES
The comparative accuracy of LUS and CXR in diagnosing ADHF as measured by the differences between the 2 modalities in pooled sensitivity and specificity.
RESULTS
The literature search yielded 1377 nonduplicate titles that were screened, of which 43 articles (3.1%) underwent full-text review. Six studies met the inclusion criteria, representing a total of 1827 patients. Pooled estimates for LUS were 0.88 (95% Cl, 0.75-0.95) for sensitivity and 0.90 (95% Cl, 0.88-0.92) for specificity. Pooled estimates for CXR were 0.73 (95% CI, 0.70-0.76) for sensitivity and 0.90 (95% CI, 0.75-0.97) for specificity. The relative sensitivity ratio of LUS, compared with CXR, was 1.2 (95% CI, 1.08-1.34; P < .001), but no difference was found in specificity between tests (relative specificity ratio, 1.0; 95% CI, 0.90-1.11; P = .96).
CONCLUSIONS AND RELEVANCE
The findings suggest that LUS is more sensitive than CXR in detecting pulmonary edema in ADHF; LUS should be considered as an adjunct imaging modality in the evaluation of patients with dyspnea at risk of ADHF.
Topics: Aged; Aged, 80 and over; Female; Heart Failure; Humans; Lung; Male; Middle Aged; Point-of-Care Systems; Predictive Value of Tests; Radiography, Thoracic; Ultrasonography
PubMed: 30874784
DOI: 10.1001/jamanetworkopen.2019.0703