AI‐enhanced micro‐ultrasound improves detection of clinically significant prostate cancer at biopsy

PurposeThe purpose of the study was to evaluate the diagnostic significance of two new and a few clinical markers for prostate cancer (PCa) at various prostate volumes (PV).MethodsThe study subjects were divided into two groups. Among them, there were 70 cases in the PV ≤30 ml group (benign prostatic hyperplasia [BPH]: 32 cases, PCa: 38 cases) and 372 cases in the PV > 30 ml group (BPH: 277 cases, PCa: 95 cases). SPSS 26.0 and GraphPad Prism 8.0 were used to construct their receiver operating characteristic (ROC) curves for diagnosing PCa and calculating their area under the ROC curve (AUC).ResultsIn the PV ≤30 ml group, the diagnostic parameters based on prostate‐specific antigen (PSA) had a decreased diagnostic significance for PCa. In the PV > 30 ml group, PSAD (AUC = 0.709), AVR (AVR = Age/PV, AUC = 0.742), and A‐PSAD (A‐PSAD = Age×PSA/PV, AUC = 0.736) exhibited moderate diagnostic significance for PCa, which was better than PSA‐AV (AUC = 0.672), free PSA (FPSA, AUC = 0.509), total PSA (TPSA, AUC = 0.563), (F/T) PSA (AUC = 0.540), and (F/T)/PSAD (AUC = 0.663). Compared with AVR, A‐PSAD exhibited similar diagnostic significance for PCa, but higher than PSA density (PSAD).ConclusionsChoosing appropriate indicators for different PVs could contribute to the early screening and diagnosis of PCa. The difference in the diagnostic value of two new indicators (A‐PSAD and AVR), and PSAD for PCa may require further validation by increasing the sample size.

PD39-12 FOLLOW-UP MODALITIES AFTER FOCAL THERAPY FOR CLINICALLY LOCALIZED PROSTATE CANCER: MRI, CEUS, PSA, AND BIOPSY

The Conduct and Reporting of Meta-Analyses of Studies of Diagnostic Tests, and a Consideration of ROC Curves: Answers to the January 2010 Journal Club Questions

Background: Beyond prostate-specific antigen (PSA), other biomarkers for prostate cancer (PCa) detection are available and need to be evaluated for clinical routine. Objective: The aim of the study was to evaluate the Prostate Health Index (PHI) density (PHID) in comparison with PHI in a large Caucasian group >1,000 men. Methods: PHID values were used from available patient data with PSA, free PSA, and [−2]pro­PSA and prostate volume from 3 former surveys from 2002 to 2014. Those 1,446 patients from a single-center cohort included 701 men with PCa and 745 with no PCa. All patients received initial or repeat biopsies. The diagnostic accuracy was evaluated by receiver operating characteristic (ROC) curves comparing area under the ROC curves (AUCs), precision-recall approach, and decision curve analysis (DCA). Results: PHID medians differed almost 2-fold between PCa (1.12) and no PCa (0.62) in comparison to PHI (48.6 vs. 33; p always <0.0001). However, PHID and PHI were equal regarding the AUC (0.737 vs. 0.749; p = 0.226), and the curves of the precision-recall analysis also overlapped in the sensitivity range between 70 and 100%. DCA had a maximum net benefit of only ∼5% for PHID versus PHI between 45 and 55% threshold probability. Contrary, in the 689 men with a prostate volume ≤40 cm³, PHI (AUC 0.732) showed a significant larger AUC than PHID (AUC 0.69, p = 0.014). Conclusions: Based on DCA, PHID had only a small advantage in comparison with PHI alone, while ROC analysis and precision-recall analysis showed similar results. In smaller prostates, PHI even outperformed PHID. The increment for PHID in this large Caucasian cohort is too small to justify a routine clinical use.

The objective of this study was to evaluate the prostate specific antigen (PSA) density (PSAD) (the quotient of PSA and prostate volume) compared with the percent free PSA (%fPSA) in different total PSA (tPSA) ranges from 2 ng/mL to 20 ng/mL. Possible cut-off levels depending on the tPSA should be established. In total, 1809 men with no pretreatment of the prostate were enrolled between 1996 and 2004. Total and free PSA were measured with the IMMULITE PSA and Free PSA kits (Diagnostic Products, Los Angeles, CA). Prostate volume was determined by transrectal ultrasound. The diagnostic validity of tPSA, %fPSA, and PSAD was evaluated by receiver operation characteristic (ROC) curve analysis. The PSAD differed significantly (P < 0.0001) between patients with prostate carcinoma and patients with benign prostatic hyperplasia in all analyzed ranges of tPSA and prostate volume. At the 90% and 95% sensitivity levels and regarding the area under the ROC curve (AUC) within the tPSA range of 2-4 ng/mL, The PSAD was significantly better than tPSA and %fPSA. Within the tPSA range of 4-10 ng/mL, the PSAD did not perform better than %fPSA. PSAD showed a better performance than %fPSA at tPSA concentrations < 4 ng/mL for detecting prostate carcinoma, with a significantly larger AUC for PSAD (0.739) compared with %fPSA (0.667). PSAD did not perform better than %fPSA when the tPSA range of 4-10 ng/mL was analyzed. Different PSAD cut-off values of 0.05 at tPSA 2-4 ng/mL, 0.1 at tPSA 4-10 ng/mL, and 0.19 at 10-20 ng/mL were necessary to reach 95% sensitivity.

To develop a logistic regression-based model to predict prostate cancer biopsy at, and compare its performance to the risk calculator developed by the Prostate Cancer Prevention Trial (PCPT), which was based on age, race, prostate-specific antigen (PSA) level, a digital rectal examination (DRE), family history, and history of a previous negative biopsy, and to PSA level alone. We retrospectively analysed the data of 1280 men who had a biopsy while enrolled in a prospective, multicentre clinical trial. Of these, 1108 had all relevant clinical and pathological data available, and no previous diagnosis of prostate cancer. Using the PCPT risk calculator, we calculated the risks of prostate cancer and of high-grade disease (Gleason score > or =7) for each man. Receiver operating characteristic (ROC) curves for the risk calculator, PSA level and the novel regression-based model were compared. Prostate cancer was detected in 394 (35.6%) men, and 155 (14.0%) had Gleason > or =7 disease. For cancer prediction, the area under the ROC curve (AUC) for the risk calculator was 66.7%, statistically greater than the AUC for PSA level of 61.9% (P < 0.001). For predicting high-grade disease, the AUCs were 74.1% and 70.7% for the risk calculator and PSA level, respectively (P = 0.024). The AUCs increased to 71.2% (P < 0.001) and 78.7% (P = 0.001) for detection and high-grade disease, respectively, with our novel regression-based models. ROC analyses show that the PCPT risk calculator modestly improves the performance of PSA level alone in predicting an individual's risk of prostate cancer or high-grade disease on biopsy. This predictive tool might be enhanced by including percentage free PSA and the number of biopsy cores.

Background: Prostate cancer detection is currently based on serum PSA with a digital rectal examination which is neither specific nor sensitive, which caused many unnecessary prostate gland biopsies that are highly expensive and can result in unwanted complications. Serum free PSA increases with a larger prostate gland,yet declines with a gland that contains cancer cells, thus prompting the hypothesis that calculating the ratio of serum free PSA against prostate gland volume provides the so-called “free PSA density” which can be utilized to improve prostate cancer detection. Methods: Male participants were deemed eligible if they are at risk of prostate cancer with a PSA level of 4-10 ng/dL and aged between 50-75 years. Serum PSA and serum free PSA were obtained concurrently, followed by transrectal ultrasonography for prostate volume calculation and biopsy of the gland. Also reported are the sensitivity, specificity, positive predictive value, negative predictive value, accuracy, and receiver operating characteristic(ROC) with area under the ROC curve(AUC) of serum PSA, %free PSA ratio, free PSA density and PSA density. The AUC of the variables were compared with the free PSA density and reported. Results: The free PSA density cut point values which provided the highest accuracy was 0.025mg/mL/cc, which had 61.5% sensitivity and 67.25% specificity. The ROC results indicate that %free PSA ratio had the best AUC at 0.86. Free PSA density and PSA density have AUC at 0.65 and 0.61, respectively. Meanwhile, serum PSA had the worst AUC of 0.54. The researchers also calculated different AUCs of other variables to free PSA density. Finally, the AUC of free PSA density was significantly better than the reference standard tool serum PSA (p=0.022). Conclusions: Prostate cancer is an emerging cancer among elderly men. Frequent use of serum PSA as a screening tool allows earlier diagnosis of this cancer but with the expensive of unnecessary further investigations. Most novel and promising tools are too expensive to be used as a generalized screening tool. From this study, free PSA density may be a reasonable alternative tool for detection of prostate cancer.

Purpose of study Methods of comparing the accuracy of diagnostic tests are of increasing necessity in biomedical science. When a test result is measured on a continuous scale, an assessment of the performance of the overall value of the test can be made using the Receiver Operating Characteristic (ROC) curve. This curve describes the discrimination ability of a diagnosis test in terms of diseased subjects from non-diseased subjects. The area under the ROC curve (AUC) describes the probability that a randomly chosen diseased subject will have higher probability of having disease than a randomly chosen non-diseased subject. For comparing two or more diagnostic test results, the difference between AUCs is often used. This paper proposes a non-parametric alternative method of comparing two or more correlated area under the curve (AUCs) of diagnostic tests for paired sample data. This method is based on Chi-square test statistic. Methods This paper investigated both parametric and non-parametric methods of comparing the equality of two AUCs and proposed a Chi-square test for the comparison of two or more diagnostic test processes. The proposed method does not require the knowledge of true status of subjects or gold standard in evaluating the accuracy of tests unlike other existing methods. The proposed method is most suitable for paired sample design. It also offers reliable statistical inferences even in small sample problems and circumvent the difficulties of deriving the statistical moments of complex summary statistics as seen in the Delong method. The proposed method provides for further analysis to determine the possible reason for rejecting the null hypothesis of equality of AUCs. Results The proposed method when applied on real data, avoids the lengthy and more difficult procedures of estimating the variances of two AUCs as a way of determining if two AUCs differ significantly. The method is validated using the Cochran Q test and was shown to compare favourably. The proposed method recommended for comparing two or more correlated AUCs when the data is paired. It is simple and does not require prior knowledge of true status of subjects unlike other existing methods. Keywords: Chi-square test, Cochran Q test, cut-off value, area under the curve, receiver operating characteristic, Dichotomous data DOI : 10.7176/JNSR/9-9-06 Publication date :May 31 st 2019

Outcome prediction by serum calprotectin in patients with COVID-19 in the emergency department

Objective To evaluate the performance of an artificial intelligence (AI) assisted diagnosis system for diabetic retinopathy (DR) based on deep learning theory. Methods Diagnostic performance of a robot assisted diagnosis system called SongYue for DR was trained by using 25 297 retinal images tagged by fundus doctors from multiple hospitals in China.Four types of DR detection model consisting of abnormal DR, referable DR, severe non-proliferative and proliferative DR as well as proliferative DR according to fundus leisions identification were established.The ability of the system to distinguish DR was determined by using receiver operator characteristic (ROC) analysis, sensitivity and specificity of the system. Results SongYue system achieved an area under the ROC curve (AUC) of 0.920 for successfully distinguishing normal images from those DR with a sensitivity of 96.0% at a specificity of 87.9%.The AUC of SongYue for referable DR was 0.925, sensitivity was 90.4%, and specificity was 95.2%.For severe non-proliferative and proliferative DR, AUC was 0.845, sensitivity was 72.7%, and specificity was 96.2%.For proliferative DR, AUC was 0.855, sensitivity was 73.5%, and specificity was 97.3%. Conclusions SongYue robot assisted diagnosis system has high AUC, sensitivity and specificity for identifying DR, showing good clinical applicable benefits. Key words: Diabetic retinopathy; Artificial intelligence; Robot; Assisted diagnosis; Deep learning

Pregnancies complicated by late-onset fetal growth restriction (FGR) are at increased risk of short- and long-term morbidities. Despite this, identification of cases at higher risk of adverse perinatal outcome, at the time of FGR diagnosis, is challenging. The aims of this study were to elucidate the strength of association between fetoplacental Doppler indices at the time of diagnosis of late-onset FGR and adverse perinatal outcome, and to determine their predictive accuracy. This was a prospective study of consecutive singleton pregnancies complicated by late-onset FGR. Late-onset FGR was defined as estimated fetal weight (EFW) or abdominal circumference (AC) < 3rd centile, or EFW or AC < 10th centile and umbilical artery (UA) pulsatility index (PI) > 95th centile or cerebroplacental ratio (CPR) < 5th centile, diagnosed after 32 weeks. EFW, uterine artery PI, UA-PI, fetal middle cerebral artery (MCA) PI, CPR and umbilical vein blood flow normalized for fetal abdominal circumference (UVBF/AC) were recorded at the time of the diagnosis of FGR. Doppler variables were expressed as Z-scores for gestational age. Composite adverse perinatal outcome was defined as the occurrence of at least one of emergency Cesarean section for fetal distress, 5-min Apgar score < 7, umbilical artery pH < 7.10 and neonatal admission to the special care unit. Logistic regression analysis was used to elucidate the strength of association between different ultrasound parameters and composite adverse perinatal outcome, and receiver-operating-characteristics (ROC)-curve analysis was used to determine their predictive accuracy. In total, 243 consecutive singleton pregnancies complicated by late-onset FGR were included. Composite adverse perinatal outcome occurred in 32.5% (95% CI, 26.7-38.8%) of cases. In pregnancies with composite adverse perinatal outcome, compared with those without, mean uterine artery PI Z-score (2.23 ± 1.34 vs 1.88 ± 0.89, P = 0.02) was higher, while Z-scores of UVBF/AC (-1.93 ± 0.88 vs -0.89 ± 0.94, P ≤ 0.0001), MCA-PI (-1.56 ± 0.93 vs -1.22 ± 0.84, P = 0.004) and CPR (-1.89 ± 1.12 vs -1.44 ± 1.02, P = 0.002) were lower. On multivariable logistic regression analysis, Z-scores of mean uterine artery PI (P = 0.04), CPR (P = 0.002) and UVBF/AC (P = 0.001) were associated independently with composite adverse perinatal outcome. UVBF/AC Z-score had an area under the ROC curve (AUC) of 0.723 (95% CI, 0.64-0.80) for composite adverse perinatal outcome, demonstrating better accuracy than that of mean uterine artery PI Z-score (AUC, 0.593; 95% CI, 0.50-0.69) and CPR Z-score (AUC, 0.615; 95% CI, 0.52-0.71). A multiparametric prediction model including Z-scores of MCA-PI, uterine artery PI and UVBF/AC had an AUC of 0.745 (95% CI, 0.66-0.83) for the prediction of composite adverse perinatal outcome. While CPR and uterine artery PI assessed at the time of diagnosis are associated independently with composite adverse perinatal outcome in pregnancies complicated by late-onset FGR, their diagnostic performance for composite adverse perinatal outcome is low. UVBF/AC showed better accuracy for prediction of composite adverse perinatal outcome, although its usefulness in clinical practice as a standalone predictor of adverse pregnancy outcome requires further research. Copyright © 2019 ISUOG. Published by John Wiley & Sons Ltd.

Previous studies have reported that microRNAs are implicated in the pathogenesis of diabetic nephropathy (DN). In this study, the underlying molecular mechanisms and diagnostic significance of miR-372-3p were investigated in the process of DN. Cell proliferation and apoptosis were measured using MTT and Annexin V-FITC double staining, respectively. RT-qPCR and western blotting were used to measure the expression levels of mRNA and protein. The diagnostic power of miR-372-3p in plasma for DN was evaluated using the receiver operating characteristics (ROC) curves and the area under the ROC curves (AUC). miR microarray analysis revealed that 126 miRs were significantly differentially expressed in response to high glucose stimulation. Among these miRs, high glucose stimulated miR-372-3p expression at the highest level. In vitro experimental measurements showed that knockdown of miR-372-3p showed the ability to reverse high glucose-induced glomerular endothelial cell apoptosis and impairment of eNOS/NO bioactivity. Mechanistic analysis revealed that fibroblast growth factor-16 (FGF-16) as a direct of miR-372-3p protected against high glucose-induced glomerular endothelial cell dysfunction. ROC analysis revealed that the diagnostic value of miR-372-3p, miR-15a or miR-372-3p combined with miR-15a in type 2 diabetes mellitus patients (AUC = 0.841, p < 0.001; AUC = 0.822, p < 0.001 or AUC = 0.922, p < 0.001) with DN was better than in type 1 diabetes mellitus patients (AUC = 0.805, p < 0.001; AUC = 0.722, p < 0.001 or AUC = 0.865, p < 0.001) with DN. miR-372-3p might be a valuable therapeutic target and diagnostic marker for patients with DN.

Landslide risk prevention requires the delineation of landslide-prone areas as accurately as possible. Therefore, selecting a method or a technique that is capable of providing the highest landslide prediction capability is highly important. The main objective of this study is to assess and compare the prediction capability of advanced machine learning methods for landslide susceptibility mapping in the Mila Basin (Algeria). First, a geospatial database was constructed from various sources. The database contains 1156 landslide polygons and 16 conditioning factors (altitude, slope, aspect, topographic wetness index (TWI), landforms, rainfall, lithology, stratigraphy, soil type, soil texture, landuse, depth to bedrock, bulk density, distance to faults, distance to hydrographic network, and distance to road networks). Subsequently, the database was randomly resampled into training sets and validation sets using 5 times repeated 10 k-folds cross-validations. Using the training and validation sets, five landslide susceptibility models were constructed, assessed, and compared using Random Forest (RF), Gradient Boosting Machine (GBM), Logistic Regression (LR), Artificial Neural Network (NNET), and Support Vector Machine (SVM). The prediction capability of the five landslide models was assessed and compared using the receiver operating characteristic (ROC) curve, the area under the ROC curves (AUC), overall accuracy (Acc), and kappa index. Additionally, Wilcoxon signed-rank tests were performed to confirm statistical significance in the differences among the five machine learning models employed in this study. The result showed that the GBM model has the highest prediction capability (AUC = 0.8967), followed by the RF model (AUC = 0.8957), the NNET model (AUC = 0.8882), the SVM model (AUC = 0.8818), and the LR model (AUC = 0.8575). Therefore, we concluded that GBM and RF are the most suitable for this study area and should be used to produce landslide susceptibility maps. These maps as a technical framework are used to develop countermeasures and regulatory policies to minimize landslide damages in the Mila Basin. This research demonstrated the benefit of selecting the best-advanced machine learning method for landslide susceptibility assessment.

Актуальность. Новые возможности улучшения функции внешнего дыхания (ФВД) связаны с применением высокочастотной осцилляции грудной клетки (ВЧОГК), которая позволяет восстановить дренаж бронхиального дерева и оптимизировать легочную вентиляцию. Цель исследования — изучение и оценка ФВД у детей с внебольничной пневмонией (ВП). Материалы и методы. Проведено обследование 107 детей (основная группа (ОГ) — 55 человек и контрольная группа (КГ) — 52 человека) в возрасте 6–17 лет с ВП с острым и неосложненным течением. Дети ОГ получали базисную терапию (БТ) с проведением процедур ВЧОГК, тогда как дети КГ получали исключительно БТ. Исследовали ФВД в динамике лечения, применяя спирометрию. Результаты. В начале терапии ВП показатели ФВД в исследуемых группах имели несущественные различия. При анализе параметров ФВД у детей ОГ на 10-й день БТ отмечено повышение объема форсированного выдоха за 1-ю секунду (ОФВ1) (88,36 ± 1,55 %, p = 0,02), жизненной емкости легких (ЖЕЛ) (88,18 ± 1,53 %, p = 0,02), форсированной жизненной емкости легких (ФЖЕЛ) (86,77 ± 1,37 %, p = 0,03), максимальной объемной скорости воздуха на уровне выдоха 25 % (МОШ25) (90,10 ± 2,99 %, p = 0,02) и максимальной вентиляции легких (МВЛ) (88,31 ± 1,70 %, p = 0,04). Сравнивая показатели ФВД детей КГ, выявили менее выраженную динамику, в частности, ОФВ1 (81,65 ± 2,44 %), ЖЕЛ (82,95 ± 2,56), ФЖЕЛ (80,85 ± 2,09), МОШ25 (82,63 ± 3,08 %) и МВЛ (85,65 ± 1,99 %). При проведении ROC-анализа динамика восстановления ФВД у детей ОГ более заметна за счет улучшения МВЛ, о чем свидетельствует самая большая площадь под кривой ROC (ППК) — 0,99, ОФВ1 — ППК 0,94 и ОФВ1/ФЖЕЛ — ППК 0,94. У детей КГ была менее выражена динамика исследуемых параметров ФВД, что подтверждается МВЛ — ППК 0,63, ОФВ1 — ППК 0,79 и ОФВ1/ФЖЕЛ — ППК 0,89. Выводы. Для улучшения вентиляционной функции легких у детей с ВП необходимо проводить процедуры бронходренажа, используя ВЧОГК в составе БТ, о чем свидетельствуют полученные положительные результаты спирометрии.

Background:Whether the Glasgow Coma Scale (GCS) can assess intubated patients is still a topic of controversy. We compared the test performance of the GCS motor component (GCS-M)/Simplified Motor Score (SMS) to the total of the GCS in predicting the outcomes of intubated acute severe cerebral vascular disease patients.Methods:A retrospective analysis of prospectively collected observational data was performed. Between January 2012 and October 2015, 106 consecutive acute severe cerebral vascular disease patients with intubation were included in the study. GCS, GCS-M, GCS eye-opening component, and SMS were documented on admission and at 24, 48, and 72 h after admission to Neurointensive Care Unit (NCU). Outcomes were death and unfavorable prognosis (modified Rankin Scale: 5–6) at NCU discharge. The receiver operating characteristic (ROC) curve was obtained to determine the prognostic performance and best cutoff value for each scoring system. Comparison of the area under the ROC curves (AUCs) was performed using the Z-test.Results:Of 106 patients included in the study, 41 (38.7%) patients died, and 69 (65.1%) patients had poor prognosis when discharged from NCU. The four time points within 72 h of admission to the NCU were equivalent for each scale's predictive power, except that 0 h was the best for each scale in predicting outcomes of patients with right-hemisphere lesions. Nonsignificant difference was found between GCS-M AUCs and GCS AUCs in predicting death at 0 h (0.721 vs. 0.717, Z = 0.135, P = 0.893) and 72 h (0.730 vs. 0.765, Z = 1.887, P = 0.060), in predicting poor prognosis at 0 h (0.827 vs. 0.819, Z = 0.395, P = 0.693), 24 h (0.771 vs. 0.760, Z = 0.944, P = 0.345), 48 h (0.732 vs. 0.741, Z = 0.593, P = 0.590), and 72 h (0.775 vs. 0.780, Z = 0.302, P = 0.763). AUCs in predicting death for patients with left-hemisphere lesions ranged from 0.700 to 0.804 for GCS-M and from 0.700 to 0.824 for GCS, in predicting poor prognosis ranged from 0.841 to 0.969 for GCS-M and from 0.875 to 0.969 for GCS, with no significant difference between GCS-M AUCs and GCS AUCs within 72 h (P > 0.05). No significant difference between GCS-M AUCs and GCS AUCs was found in predicting death (0.964 vs. 0.964, P = 1.000) and poor prognosis (1.000 vs. 1.000, P = 1.000) for patients with right-hemisphere lesions at 0 h. AUCs in predicting death for patients with brainstem or cerebella were poor for GCS-M (<0.700), in predicting poor prognosis ranged from 0.727 to 0.801 for GCS-M and from 0.704 to 0.820 for GCS, with no significant difference between GCS-M AUCs and GCS AUCs within 72 h (P > 0.05). The SMS AUCs (<0.700) in predicting outcomes were poor.Conclusions:The GCS-M approaches the same test performance as the GCS in assessing the prognosis of intubated acute severe cerebral vascular disease patients. The GCS-M could be accurately and reliably applied in patients with hemisphere lesions, but caution must be taken for patients with brainstem or cerebella lesions.