Comparisons of deep learning algorithms for diagnosing bacterial keratitis via external eye photographs

AbstractBacterial keratitis (BK), a painful and fulminant bacterial infection of the cornea, is the most common type of vision-threatening infectious keratitis (IK). A rapid clinical diagnosis by an ophthalmologist may often help prevent BK patients from progression to corneal melting or even perforation, but many rural areas cannot afford an ophthalmologist. Thanks to the rapid development of deep learning (DL) algorithms, artificial intelligence via image could provide an immediate screening and recommendation for patients with red and painful eyes. Therefore, this study aims to elucidate the potentials of different DL algorithms for diagnosing BK via external eye photos. External eye photos of clinically suspected IK were consecutively collected from five referral centers. The candidate DL frameworks, including ResNet50, ResNeXt50, DenseNet121, SE-ResNet50, EfficientNets B0, B1, B2, and B3, were trained to recognize BK from the photo toward the target with the greatest area under the receiver operating characteristic curve (AUROC). Via five-cross validation, EfficientNet B3 showed the most excellent average AUROC, in which the average percentage of sensitivity, specificity, positive predictive value, and negative predictive value was 74, 64, 77, and 61. There was no statistical difference in diagnostic accuracy and AUROC between any two of these DL frameworks. The diagnostic accuracy of these models (ranged from 69 to 72%) is comparable to that of the ophthalmologist (66% to 74%). Therefore, all these models are promising tools for diagnosing BK in first-line medical care units without ophthalmologists.

Download Full-text

Diagnostic accuracy of a urine dipstick for detecting albuminuria in hypertensive patients

F1000Research ◽

10.12688/f1000research.25564.3 ◽

2021 ◽

Vol 9 ◽

pp. 1244

Author(s):

Phornwipa Panta ◽

Win Techakehakij

Keyword(s):

Diagnostic Accuracy ◽

Predictive Value ◽

Operating Characteristic ◽

Characteristic Curve ◽

Hypertensive Patients ◽

Spot Urine ◽

Urine Dipstick ◽

Urine Albumin ◽

Low Sensitivity ◽

Sensitivity Specificity

Background: Screening for albuminuria is generally recommended among patients with hypertension. While the urine dipstick is commonly used for screening urine albumin, there is little evidence about its diagnostic accuracy among these patients in Thailand. This study aimed to assess the diagnostic accuracy of a dipstick in Thai hypertensive patients for detecting albuminuria. Methods: This study collected the data of 3,067 hypertensive patients, with the results of urine dipstick and urine albumin-to-creatinine ratio (ACR) from random single spot urine being examined in the same day at least once, at Lampang Hospital, Thailand, during 2018. For ACR, a reference standard of ≥ 30 mg/g was applied to indicate the presence of albuminuria. Results: The sensitivity, specificity, positive predictive value (PPV), and negative predictive value of the trace result from dipsticks were 53.6%, 94.5%, 86.5%, and 75.5%, respectively. The area under the receiver operating characteristic curve of the dipstick was 0.748. Conclusion: Using the dipstick for screening albuminuria among hypertensive patients should not be recommended for mass screening due to its low sensitivity. In response to high PPV, a trace threshold of the dipstick may be used to indicate presence of albuminuria.

Download Full-text

Diagnostic accuracy of a urine dipstick for detecting albuminuria in hypertensive patients

F1000Research ◽

10.12688/f1000research.25564.1 ◽

2020 ◽

Vol 9 ◽

pp. 1244

Author(s):

Phornwipa Panta ◽

Win Techakehakij

Keyword(s):

Diagnostic Accuracy ◽

Predictive Value ◽

Operating Characteristic ◽

Characteristic Curve ◽

Hypertensive Patients ◽

Spot Urine ◽

Urine Dipstick ◽

Urine Albumin ◽

Low Sensitivity ◽

Sensitivity Specificity

Background: Screening for albuminuria is generally recommended among patients with hypertension. While the urine dipstick is commonly used for screening urine albumin, there is little evidence about its diagnostic accuracy among these patients. This study aimed to assess the diagnostic accuracy of a dipstick in Thai hypertensive patients for detecting albuminuria. Methods: This study collected the data of 3,067 hypertensive patients, with the results of urine dipstick and urine albumin-to-creatinine ratio (ACR) from random single spot urine being examined in the same day at least once, at Lampang Hospital, Thailand, during 2018. For ACR, a reference standard of ≥ 30 mg/g was applied to indicate the presence of albuminuria. Results: The sensitivity, specificity, positive predictive value (PPV), and negative predictive value of the trace result from dipsticks were 53.6%, 94.5%, 86.5%, and 75.5%, respectively. The area under the receiver operating characteristic curve of the dipstick was 0.748. Conclusion: Using the dipstick for screening albuminuria among hypertensive patients should not be recommended due to its low sensitivity. In response to high PPV, a trace threshold of the dipstick may be used to indicate presence of albuminuria.

Download Full-text

The value of S-Detect in improving the diagnostic performance of radiologists for the differential diagnosis of thyroid nodules

Medical Ultrasonography ◽

10.11152/mu-2501 ◽

2020 ◽

Vol 22 (4) ◽

pp. 415

Author(s):

Qi Wei ◽

Shu-E Zeng ◽

Li-Ping Wang ◽

Yu-Jing Yan ◽

Ting Wang ◽

...

Keyword(s):

Diagnostic Accuracy ◽

Predictive Value ◽

Diagnostic Performance ◽

Thyroid Nodules ◽

Characteristic Curve ◽

Diagnostic Value ◽

Ultrasound Images ◽

Experienced Radiologist ◽

Before And After ◽

Sensitivity Specificity

Aims: To compare the diagnostic value of S-Detect (a computer aided diagnosis system using deep learning) in differentiating thyroid nodules in radiologists with different experience and to assess if S-Detect can improve the diagnostic performance of radiologists.Materials and methods: Between February 2018 and October 2019, 204 thyroid nodules in 181 patients were included. An experienced radiologist performed ultrasound for thyroid nodules and obtained the result of S-Detect. Four radiologists with different experience on thyroid ultrasound (Radiologist 1, 2, 3, 4 with 1, 4, 9, 20 years, respectively) analyzed the conventional ultrasound images of each thyroid nodule and made a diagnosis of “benign” or “malignant” based on the TI-RADS category. After referring to S-Detect results, they re-evaluated the diagnoses. The diagnostic performance of radiologists was analyzed before and after referring to the results of S-Detect.Results: The accuracy, sensitivity, specificity, positive predictive value and negative predictive value of S-Detect were 77.0, 91.3, 65.2, 68.3 and 90.1%, respectively. In comparison with the less experienced radiologists (radiologist 1 and 2), S-Detect had a higher area under receiver operating characteristic curve (AUC), accuracy and specificity (p <0.05). In comparison with the most experienced radiologist, the diagnostic accuracy and AUC were lower (p<0.05). In the less experienced radiologists, the diagnostic accuracy, specificity and AUC were significantly improved when combined with S-Detect (p<0.05), but not for experienced radiologists (radiologist 3 and 4) (p>0.05).Conclusions: S-Detect may become an additional diagnostic method for the diagnosis of thyroid nodules and improve the diagnostic performance of less experienced radiologists.

Download Full-text

Diagnostic accuracy of a urine dipstick for detecting albuminuria in hypertensive patients

F1000Research ◽

10.12688/f1000research.25564.2 ◽

2021 ◽

Vol 9 ◽

pp. 1244

Author(s):

Phornwipa Panta ◽

Win Techakehakij

Keyword(s):

Diagnostic Accuracy ◽

Predictive Value ◽

Operating Characteristic ◽

Characteristic Curve ◽

Hypertensive Patients ◽

Spot Urine ◽

Urine Dipstick ◽

Urine Albumin ◽

Low Sensitivity ◽

Sensitivity Specificity

Download Full-text

Noninvasive Evaluation of Coronary Artery Bypass Grafts with 16-Slice Multidetector Computed Tomography

The Heart Surgery Forum ◽

10.1532/hsf98.20111075 ◽

2012 ◽

Vol 15 (1) ◽

pp. 12 ◽

Cited By ~ 1

Author(s):

Levent Sahiner ◽

Ali Oto ◽

Kudret Aytemir ◽

Tuncay Hazirolan ◽

Musturay Karcaaltincaba ◽

...

Keyword(s):

Computed Tomography ◽

Coronary Artery ◽

Diagnostic Accuracy ◽

Multidetector Computed Tomography ◽

Coronary Artery Bypass ◽

Predictive Value ◽

Artery Bypass ◽

Time Interval ◽

Mdct Angiography ◽

Sensitivity Specificity

Background: The aim of this study was to investigate the diagnostic accuracy of 16-slice multislice, multidetector computed tomography (MDCT) angiography for the evaluation of grafts in patients with coronary artery bypass grafting (CABG).Methods: Fifty-eight consecutive patients with CABG who underwent both MDCT and conventional invasive coronary angiography were included. The median time interval between the 2 procedures was 10 days (range, 1-32 days). The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of MDCT for the detection of occluded grafts were calculated. The accuracy of MDCT angiography for detecting significant stenoses in patent grafts and the evaluability of proximal and distal anastomoses were also investigated.Results: Optimal diagnostic images could not be obtained for only 3 (2%) of 153 grafts. Evaluation of the remaining 150 grafts revealed values for sensitivity, specificity, PPV, NPV, and diagnostic accuracy of the MDCT angiography procedure for the diagnosis of occluded grafts of 87%, 97%, 94%, 93%, and 92%, respectively. All of the proximal anastomoses were optimally visualized. In 4 (8%) of 50 patent arterial grafts, however, the distal anastomotic region could not be evaluated because of motion and surgical-clip artifacts. The accuracy of MDCT angiography for the detection of significant stenotic lesions was relatively low (the sensitivity, specificity, PPV, and NPV were 67%, 98%, 50%, and 99%, respectively). The number of significant lesions was insufficient to reach a reliable conclusion, however.Conclusion: Our study showed that MDCT angiography with 16-slice systems has acceptable diagnostic performance for the evaluation of coronary artery bypass graft patency.

Download Full-text

The role of serum periostin in the diagnosis of asthma: A meta-analysis

Allergy and Asthma Proceedings ◽

10.2500/aap.2020.41.200038 ◽

2020 ◽

Vol 41 (4) ◽

pp. 240-247

Author(s):

Lei Yang ◽

Qingtao Zhao ◽

Shuyu Wang

Keyword(s):

Diagnostic Accuracy ◽

Meta Analysis ◽

Characteristic Curve ◽

False Negative ◽

Summary Receiver Operating Characteristic ◽

True Negative ◽

Negative Results ◽

Noninvasive Biomarker ◽

Sensitivity Specificity ◽

Serum Periostin

Background: Serum periostin has been proposed as a noninvasive biomarker for asthma diagnosis and management. However, its accuracy for the diagnosis of asthma in different populations is not completely clear. Methods: This meta-analysis aimed to evaluate the diagnostic accuracy of periostin level in the clinical determination of asthma. Several medical literature data bases were searched for relevant studies through December 1, 2019. The numbers of patients with true-positive, false-positive, false-negative, and true-negative results for the periostin level were extracted from each individual study. We assessed the risk of bias by using Quality Assessment of Diagnostic Accuracy Studies 2. We used the meta-analysis to produce summary estimates of accuracy. Results: In total, nine studies with 1757 subjects met the inclusion criteria. The pooled estimates of sensitivity, specificity, and diagnostic odds ratios for the detection of asthma were 0.58 (95% confidence interval [CI], 0.38‐0.76), 0.86 (95% CI, 0.74‐0.93), and 8.28 (95% CI, 3.67‐18.68), respectively. The area under the summary receiver operating characteristic curve was 0.82 (95% CI, 0.79‐0.85). And significant publication bias was found in this meta‐analysis (p = 0.39). Conclusion: Serum periostin may be used for the diagnosis of asthma, with moderate diagnostic accuracy.

Download Full-text

Additional value of deep learning computed tomographic angiography-based fractional flow reserve in detecting coronary stenosis and predicting outcomes

Acta Radiologica ◽

10.1177/0284185120983977 ◽

2021 ◽

pp. 028418512098397

Author(s):

Yang Li ◽

Hong Qiu ◽

Zhihui Hou ◽

Jianfeng Zheng ◽

Jianan Li ◽

...

Keyword(s):

Deep Learning ◽

Unstable Angina ◽

Fractional Flow Reserve ◽

Predictive Value ◽

Computed Tomographic Angiography ◽

Fractional Flow ◽

Computed Tomographic ◽

Flow Reserve ◽

Sensitivity Specificity ◽

Tomographic Angiography

Background Deep learning (DL) has achieved great success in medical imaging and could be utilized for the non-invasive calculation of fractional flow reserve (FFR) from coronary computed tomographic angiography (CCTA) (CT-FFR). Purpose To examine the ability of a DL-based CT-FFR in detecting hemodynamic changes of stenosis. Material and Methods This study included 73 patients (85 vessels) who were suspected of coronary artery disease (CAD) and received CCTA followed by invasive FFR measurements within 90 days. The diagnostic accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and area under the receiver operating characteristics curve (AUC) were compared between CT-FFR and CCTA. Thirty-nine patients who received drug therapy instead of revascularization were followed for up to 31 months. Major adverse cardiac events (MACE), unstable angina, and rehospitalization were evaluated and compared between the study groups. Results At the patient level, CT-FFR achieved 90.4%, 93.6%, 88.1%, 85.3%, and 94.9% in accuracy, sensitivity, specificity, PPV, and NPV, respectively. At the vessel level, CT-FFR achieved 91.8%, 93.9%, 90.4%, 86.1%, and 95.9%, respectively. CT-FFR exceeded CCTA in these measurements at both levels. The vessel-level AUC for CT-FFR also outperformed that for CCTA (0.957 vs. 0.599, P < 0.0001). Patients with CT-FFR ≤0.8 had higher rates of rehospitalization (hazard ratio [HR] 4.51, 95% confidence interval [CI] 1.08–18.9) and MACE (HR 7.26, 95% CI 0.88–59.8), as well as a lower rate of unstable angina (HR 0.46, 95% CI 0.07–2.91). Conclusion CT-FFR is superior to conventional CCTA in differentiating functional myocardial ischemia. In addition, it has the potential to differentiate prognoses of patients with CAD.

Download Full-text

Diagnostic accuracy of monofilament test to detect diabetic neuropathy

Journal of Advances in Internal Medicine ◽

10.3126/jaim.v10i1.37086 ◽

2021 ◽

Vol 10 (1) ◽

pp. 20-25

Author(s):

Sujan Shrestha ◽

Mamen Prasad Gorhaly ◽

Manil Ratna Bajracharya

Keyword(s):

Peripheral Neuropathy ◽

Positive Predictive Value ◽

Diagnostic Accuracy ◽

Negative Predictive Value ◽

Predictive Value ◽

Diabetic Peripheral Neuropathy ◽

Perception Threshold ◽

Sensitivity Specificity ◽

Vibration Perception ◽

Monofilament Test

Background Diabetic peripheral neuropathy (DPN) is a significant independent risk factor for diabetic foot, and an effective screening instrument is required to diagnose DPN early to prevent future ulceration and amputation. This study aims to determine the diagnostic accuracy of monofilament test to detect diabetic peripheral neuropathy. Methods This cross-sectional study was conducted in National Academy of Medical Sciences, Bir hospital, Mahabouddha, Kathmandu from February 2016 to January 2017. A total of 96 diabetic patients attending inpatient and outpatient Department were selected. Diabetic peripheral neuropathy was assessed by measurement of loss of protective sensation (LOPS) by monofilament test and compared with vibration perception threshold by standard biothesiometer. The sensitivity, specificity, positive predictive value and negative predictive value of monofilament test were calculated. Results The prevalence of diabetic peripheral neuropathy was 26%. The sensitivity, specificity, positive predictive value and negative predictive value of monofilament test were found to be 92.0%, 95.8%, 88.5% and 97.1% respectively. There was strong association between LOPS by monofilament and vibration perception threshold by biothesiometer. Conclusion This study showed a strong diagnostic accuracy of monofilament test to detect DPN when compared with biothesiometer. As monofilament test is a cheap, easily available, and portable, it can be used in the periphery where biothesiometer is not available.

Download Full-text

THE SENSITIVITY, SPECIFICITY AND ACCURACY OF STATIC METHOD OF ULTRASONOGRAPHY TO CONFIRM ENDOTRACHEAL INTUBATION IN EMERGENCY DEPARTMENT.

10.36106/ijsr/6729269 ◽

2021 ◽

pp. 8-13

Author(s):

Pankaj Kumar Singh ◽

Budhaditya Sanyal ◽

Mohit Bhatnagar ◽

Mandeep Joshi ◽

Shreya Verma

Keyword(s):

Diagnostic Accuracy ◽

Endotracheal Tube ◽

Predictive Value ◽

Static Method ◽

Tube Placement ◽

Short Course ◽

Endotracheal Tube Placement ◽

Linear Probe ◽

Post Intubation ◽

Sensitivity Specificity

Aims and objectives: This study aims to assess the diagnostic accuracy and timeliness of ultrasonography by static method only for identication of Endotracheal tube (ET Tube) placement in the trachea in emergency settings vs existing clinical methods. Material and Methods: This prospective study was carried out in the emergency room from October 2018 till the end of March 2019. The ultrasonography was performed in 120 emergency patients only after the intubation had been completed ie, static phase. A linear probe was used over the neck to identify the predened signs of ET intubation. Residents who perform ultrasound examination ll a form after assessment of each patient. Results: It was found that Tracheal Intubation-USG Sensitivity was 99.1, Specicity was 91.7, Positive Predictive Value: 99.1, Negative Predictive Value was 91.7 and Accuracy was 98.3%. Ultrasonography can be used as an adjunct tool to verify the ETTposition by Emergency Physicians which can be performed easily after a brieng or short-course training.Conclusion:This study demonstrates that US imaging has a high diagnostic accuracy to immediately conrm proper ETT placement post-intubation in an emergency setup. Therefore, it seems that ultrasonography using a static technique only is a proper screening tool in determining endotracheal tube placement.

Download Full-text

DIAGNOSTIC ACCURACY OF GRAY SCALE ULTRASONOGRAPHY VERSUS COLOR DOPPLER IN SUSPECTED CASES OF ACUTE APPENDICITIS

The International Journal of Frontier Sciences ◽

10.37978/tijfs.v5i1.319 ◽

2021 ◽

Vol 5 (1) ◽

Author(s):

Tahir Iqbal ◽

Muhammad Usman Shahid ◽

Ishfaq Ahmad Shad ◽

Shahzad Karim Bhatti ◽

Syed Amir Gilani ◽

...

Keyword(s):

Acute Appendicitis ◽

Positive Predictive Value ◽

Diagnostic Accuracy ◽

Negative Predictive Value ◽

Predictive Value ◽

Color Doppler ◽

Diagnostic Tools ◽

Gray Scale ◽

Versus Color ◽

Sensitivity Specificity

ABSTRACT: BACKGROUND: A common surgical emergency is acute appendicitis. Various diagnostic tools are available to diagnosis acute appendicitis. Radiological investigations play an important role in making accurate and early diagnosis and thus preventing morbidity associated with the disease. OBJECTIVE: To determine the diagnostic accuracy of gray scale ultrasonography versus color Doppler in suspected cases of acute appendicitis. MATERIALS AND METHODS: The study was carried in the department of Radiology of Mayo Hospital, Lahore. A total of 75 patients were enrolled of age 18-40 years, both genders who were suspected cases of acute appendicitis. All patients underwent baseline investigations along with gray scale ultrasonography and color Doppler. All patients were subjected to surgery to confirm the diagnosis and findings were subjected to statistical analysis. RESULTS: The mean age of the patients was 23.25 ±10.55 and mean transverse diameter of appendix was 8.37 ±3.39. There were 62.7% males and 37.3%females. Findings of gray scale ultrasonography and color Doppler were then correlated with surgical findings to calculate the diagnostic accuracy of these modalities. The results revealed that gray scale ultrasonography sensitivity, specificity, positive predictive value, negative predictive value and accuracy was 92.7%, 94.32%, 95%, 91.4% and 93.3% respectively, whereas color Doppler had sensitivity, specificity, positive predictive value, negative predictive value and accuracy of 97.7%, 93.9%, 95.3%, 97% and 96% respectively. Diagnostic accuracy of both modalities together was 98.6%. CONCLUSION: Color Doppler has better diagnostic accuracy than gray scale ultrasonography for diagnosis of acute appendicitis and the combination of both modalities yields diagnostic accuracy that is similar to gold standard.

Download Full-text