As an effective lesion heterogeneity depiction, texture information extracted from computed tomography has become increasingly important in polyp classification. However, variation and redundancy ...among multiple texture descriptors render a challenging task of integrating them into a general characterization. Considering these two problems, this work proposes an adaptive learning model to integrate multi-scale texture features.
To mitigate feature variation, the whole feature set is geometrically split into several independent subsets that are ranked by a learning evaluation measure after preliminary classifications. To reduce feature redundancy, a bottom-up hierarchical learning framework is proposed to ensure monotonic increase of classification performance while integrating these ranked sets selectively. Two types of classifiers, traditional (random forest + support vector machine)- and convolutional neural network (CNN)-based, are employed to perform the polyp classification under the proposed framework with extended Haralick measures and gray-level co-occurrence matrix (GLCM) as inputs, respectively. Experimental results are based on a retrospective dataset of 63 polyp masses (defined as greater than 3 cm in largest diameter), including 32 adenocarcinomas and 31 benign adenomas, from adult patients undergoing first-time computed tomography colonography and who had corresponding histopathology of the detected masses.
We evaluate the performance of the proposed models by the area under the curve (AUC) of the receiver operating characteristic curve. The proposed models show encouraging performances of an AUC score of 0.925 with the traditional classification method and an AUC score of 0.902 with CNN. The proposed adaptive learning framework significantly outperforms nine well-established classification methods, including six traditional methods and three deep learning ones with a large margin.
The proposed adaptive learning model can combat the challenges of feature variation through a multiscale grouping of feature inputs, and the feature redundancy through a hierarchal sorting of these feature groups. The improved classification performance against comparative models demonstrated the feasibility and utility of this adaptive learning procedure for feature integration.
Whether cigarette smoking is associated with the progression of immunoglobulin A nephropathy (IgAN) remains uncertain; therefore, we aimed to evaluate the effect of cigarette smoking on the prognosis ...of IgAN.
We divided 1239 IgAN patients from West China Hospital of Sichuan University who met the inclusion criteria into smoker (current or former) and non-smoker groups. The endpoint was end-stage renal disease (ESRD: eGFR < 15 mL/min/1.73 m
or undergoing renal replacement treatment) and/or eGFR decreased by > 50%. Kaplan-Meier, correlation, logistic regression and Cox proportional hazards analyses were performed. The association between cigarette smoking and IgAN was further verified by propensity-score-matched cohort analysis.
During the mean follow-up period of 61 months, 19% (40/209) of the smoker group and 11% (110/1030) of the non-smoker group reached the study endpoint (p < 0.001). Multivariate Cox regression analysis revealed that cigarette smoking (hazard ratio (HR) = 1.58; p = 0.043) was an independent risk factor predicting poor renal progression in IgAN, and that IgAN patients with chronic kidney disease (CKD) stage 3-4 were more susceptible to cigarette smoking (p < 0.001). After propensity score matching (PSM), a significant correlation between cigarette smoking and renal outcomes in IgAN patients was seen. Furthermore, Spearman's correlation test revealed that smoking dose was negatively correlated with eGFR (r = 0.141; p < 0.001) and positively related with proteinuria (r = 0.096; p = 0.001).
Cigarette smoking is an independent risk factor for IgAN progression, especially for advanced patients.
Background
The triglycerides to high-density lipoprotein cholesterol (TG/HDL-C) ratio is an easy-to-use atherogenic and prognostic marker which has attracted increasing attention these days. However, ...whether TG/HDL-C correlate with outcomes in IgA nephropathy (IgAN) patients remains unknown. To clarify these issues, we conducted this study.
Methods
A total of 1146 patients from West China Hospital of Sichuan University were retrospectively analysed between 2008 and 2018.The demographic, clinical and pathological data of all patients at the time of biopsy were collected. Then, patients were divided into the high TG/HDL group (TG/HDL ≥ 1.495, N=382) and the low TG/HDL group (TG/HDL-C < 1.495, N=764) based on the optimal cut-off value of the TG/HDL-C using receive operating curve. Cox proportional hazard models and Kaplan–Meier curves were used to evaluate the renal outcomes of IgAN.
Results
The median age of the patients was 33 (26-42) years, and 44.5% were men. By correlation analysis, we found that the TG/HDL-C ratio was negatively correlated with the eGFR (r = 0.250,
P
< 0.001) but positively correlated with proteinuria (r = 0.230,
P
< 0.001), BMI (r=0.380, P<0.001) and serum uric (r =0.308,
P
< 0.001). Patients with a higher TG/HDL-C ratio tended to have hypertension odds ratio (OR), 1.987; 95% CI, 1.527-2.587;
P
<0.001 and more severe pathologic lesions with tubular atrophy/interstitial fibrosis (OR, 1.610; 95% CI, 1.203-2.154;
P
=0.001). During a median follow-up period of 54.1 (35.6-73.2) months, a high TG/HDL ratio was strongly associated with worse renal survival in IgAN patients (log-rank:
P <
0.001). Multivariate Cox analysis demonstrated that a high TG/HDL-C ratio (HR 1.775, 95% CI 1.056-2.798;
P
=0.029) was an independent predictive marker to ESRD.
Conclusion
In this study, we addressed the importance of TG/HDL-C ratio as a predictive marker for IgAN progression.
The outbreak of coronavirus disease 2019 (COVID-19) has attracted global attention. During the lockdown period of COVID-19, follow-up of many patients with chronic disease had been interrupted, which ...brought severe challenges to better management of their disease. This study aimed at exploring the change of illness, daily life, and psychological responses during the COVID-19 pandemic among chronic kidney disease (CKD) patients.
A total of 612 patients were enrolled in this study; 282 patients were categorized into the CKD stage 1-2 group and 330 patients were categorized into the CKD stage 3-5 group. Among two groups, 168 (27.5%) and 177 (28.9%) patients were female with a median age of 42 and 45, respectively. The study was conducted by collecting the questionnaires in five nephrology centers. The questionnaire consisted of assessment of anxiety by using the Self-Rating Anxiety Scale and the influences of COVID-19, which included basic demographic data, the influences of COVID-19 on illness and daily life, as well as the patients' psychological responses during the epidemic.
A total of 612 patients were included and divided into two groups according to eGFR. Ninety-six patients (34%) in the CKD stage 1-2 group and 141 patients (42.7%) in the CKD stage 3-5 group had reduced their follow-up frequency (
= 0.031). More patients with CKD stages 1-2 consulted online (25.9%),
= 0.005. Besides, patients in the CKD stage 3-5 group tended to be more anxious about follow-up (
= 0.002), fearful of being infected with COVID-19 (
= 0.009), and more likely to feel symptoms getting worse (
= 0.006). The standard scores of SAS were 48.58 ± 7.082 and 51.19 ± 5.944 in the CKD stage 1-2 group and the CKD stage 3-5 group, respectively (
< 0.001). There were significant differences in the severity of anxiety (
= 0.004).
COVID-19 had a greater impact on patients with CKD stages 3-5 than those with stages 1-2 in terms of illness, daily life, and psychological disorder. Patients with CKD stages 3-5 were more anxious during the COVID-19 pandemic.
Nephronophthisis (NPHP) is a rare autosomal recessive inherited kidney disease that can cause cystic enlargement of the kidneys, and lead to end-stage renal disease (ESRD) before the age of 30 years. ...Herein we describe a case of adolescent-onset NPHP with a novel homozygous mutation in the inversin gene (
).
The patient was a 15-year-old Chinese boy who presented with ESRD. Genetic testing was performed via whole exome sequencing and validated
Sanger sequencing. A novel homozygous
mutation was identified (c. 1909C > T; p. Gln637Ter).
The results of laboratory examinations included urinary protein 1.05 g/24 h, urine erythrocyte count 5/high-power field, serum creatinine 1,026.2 μmol/L, and estimated glomerular filtration rate 5.8 ml/min/1.73 mm
. Extrarenal features included hypertension and moderate anemia, and his parents were consanguineous (first cousins). A homozygous 1-bp substitution resulting in a nonsense mutation (c. 1909C > T; p. Gln637Ter) in exon 15 of
was detected via whole exome sequencing, and validated via Sanger sequencing. According to the classification system of the American College of Medical Genetics and Genomics, the mutated gene in
is strongly pathogenic (PVS1+PM2+
P3+
P5). His parents and a younger brother were heterozygous carriers. Based on the above results he was diagnosed with juvenile type 2 NPHP. He underwent hemodialysis, and received a kidney transplant after 2 months. He is currently recovering well, with a serum creatinine level of 117 μmol/L and an estimated glomerular filtration rate of 79.6 ml/min/1.73 mm
.
Here we have described an extremely rare case of adolescent-onset type 2 NPHP caused by a homozygous
mutation. The patient had progressed to ESRD by the age of 15 years. The current report will deepen our understanding of the clinical and genetic basis of this disease.
Background: Henoch-Schönlein purpura nephritis (HSPN) is a common vasculitis involving the kidneys, with a lower incidence in adults. Meanwhile, nephrotic syndrome (NS) can appear in HSPN. However, ...the clinicopathological features and renal outcome of adult-onset HSPN presenting with NS (NS-HSPN) have not been well clarified. Methods: A total of 191 HSPN patients were prospectively analyzed and comparisons were made between NS-HSPN and non-NS-HSPN. Multivariate Cox regression analysis was carried out to find the unfavorable factors of renal outcome of NS-HSPN. Results: Among the 191 patients, 44 (23.0%) had NS-HSPN. Apart from edema and abdominal pain, patients with NS-HSPN tended to have lower levels of erythrocytes and hemoglobulin in blood as well as a greater number of erythrocytes in urine (p < 0.05). Mesangial proliferation was the most common pathological lesion in HSPN and the rates of crescent formation were significantly different, with 54.5% in NS-HSPN and 33.3% in non-NS-HSPN (p < 0.05). Notably, 18.2 and 4.8% of patients reached the composite endpoints in the NS-HSPN and non-NS-HSPN groups, respectively (p < 0.05), demonstrating that NS-HSPN patients were more likely to progress to end-stage renal disease and had a worse outcome. We also found that hypertension, estimated glomerular filtration rate (eGFR), cystatin, and tubular atrophy/interstitial fibrosis (HR > 1, p < 0.05) at onset were correlated with adverse outcome in NS-HSPN. Conclusion: NS-HSPN had more severe clinicopathological manifestations and poorer prognosis. The adverse predictors of NS-HSPN principally depend on clinicopathological presentation rather than on different therapies, and hypertension, eGFR, cystatin, and tubular atrophy/interstitial fibrosis can serve as independent risk factors in NS-HSPN.
The association of heart rate (HR) dipping pattern with renal outcomes in chronic kidney disease (CKD) patients with hypertension has never been investigated. In order to demonstrate if HR dipping ...pattern is a risk factor for renal outcomes, cardiovascular (CV) diseases, and mortality in hypertensive patients with CKD, we conducted the prospective longitudinal observational study. Patients were divided into three groups according to their nocturnal HR: HR dippers (night–day HR ratio ≤ 0.9), HR non‐dippers (0.9 < night–day HR ratio ≤ 1.0), and HR risers (night–day HR ratio > 1.0). The primary outcome was renal endpoint, a composite outcome of progression to end‐stage renal disease (ESRD) or estimated glomerular filtration rate (eGFR) decline ≥ 50%; the secondary outcomes included poor renal outcomes, CV events, and death. A total of 34 (11.3%) patients reached renal endpoint after a follow‐up of 34 ± 17 months. Both HR non‐dippers and HR risers were predictive to renal endpoint (hazard ratio 2.58, 95% confidence interval (CI) 1.04‐ 6.4, P = .04; hazard ratio 3.95, 95% CI 1.33‐ 11.79, P = .01, respectively), while only HR risers was shown to be correlated with a decline in eGFR≥ 50% (hazard ratio 5.28, 95% CI 1.45–19.16, P < .05), and decline in eGFR (β ‐0.17, 95% CI ‐0.33‐ ‐0.01, P = .04). No predictive value was found for HR dipping pattern to mortality and CV events. In conclusion, our study provided the first evidence that HR non‐dippers, especially risers were a risk factor for poor renal outcomes in hypertensive patients with CKD.
Background: The triglyceride−glucose (TyG) index is a simple, novel and reliable surrogate marker of insulin resistance. However, evidence for the prognostic impact of an elevated TyG index on IgA ...nephropathy (IgAN) is limited. Therefore, we evaluated the relationship between the TyG index and the risk of renal progression in IgAN. Method: This cohort study involved biopsy-proven IgAN between January 2009 and December 2018 in West China Hospital, in which patients were assigned to two groups based on the cut-off value of TyG using receiver operating characteristic (ROC) curves. A 1:1 matched-pair analysis was established to optimize the bias in IgAN by propensity score matching (PSM). The TyG index was calculated as ln fasting triglyceride (mg/dL) × fasting glucose (mg/dL)/2. The composite endpoint was defined by eGFR decreased ≥50% of the baseline level, end-stage kidney disease (ESKD), renal transplantation and/or death. Univariable and multivariable Cox proportional hazard models were applied to confirm the predictive value of the optimal marker. Results: Before PSM, a total of 1210 participants were ultimately included. During a median follow-up period of 55.8 months (range 37.20−79.09 months), 129 participants progressed to the composite endpoint (10.7%). After PSM, 366 patients were enrolled in the matched cohort, of whom 34 (9.3%) patients reached the endpoints. Based on the cut-off value of the TyG index, patients were divided into the low TyG index group (TyG ≤ 8.72, n = 690) and the high TyG index group (TyG > 8.72, n = 520). Further analysis demonstrated that a higher TyG index was significantly associated with a higher risk of reaching composite endpoints in IgAN patients in both the unmatched and matched cohorts (before PSM: HR 2.509, 95% CI 1.396−4.511, p = 0.002; after PSM: HR 2.654, 95% CI 1.299−5.423, p = 0.007). Conclusion: A high TyG index is associated with a higher risk of renal progression.
Computer aided detection (CADe) of pulmonary nodules plays an important role in assisting radiologists’ diagnosis and alleviating interpretation burden for lung cancer. Current CADe systems, aiming ...at simulating radiologists’ examination procedure, are built upon computer tomography (CT) images with feature extraction for detection and diagnosis. Human visual perception in CT image is reconstructed from sinogram, which is the original raw data acquired from CT scanner. In this work, different from the conventional image based CADe system, we propose a novel sinogram based CADe system in which the full projection information is used to explore additional effective features of nodules in the sinogram domain. Facing the challenges of limited research in this concept and unknown effective features in the sinogram domain, we design a new CADe system that utilizes the self-learning power of the convolutional neural network to learn and extract effective features from sinogram. The proposed system was validated on 208 patient cases from the publicly available online Lung Image Database Consortium database, with each case having at least one juxtapleural nodule annotation. Experimental results demonstrated that our proposed method obtained a value of 0.91 of the area under the curve (AUC) of receiver operating characteristic based on sinogram alone, comparing to 0.89 based on CT image alone. Moreover, a combination of sinogram and CT image could further improve the value of AUC to 0.92. This study indicates that pulmonary nodule detection in the sinogram domain is feasible with deep learning.
Background:
Bilirubin has been identified as an endogenous antioxidant and cellular protectant. The present study was performed to clarify the potential influence of serum bilirubin on IgA vasculitis ...with nephritis (IgAV-N).
Methods:
One hundred and eighty-nine IgAV-N patients over 14 years old were enrolled. The patients were divided into two groups by the optimum cut-off value calculated by ROC curve. The composite endpoints were defined as a 60% decline in estimate glomerular filtration rate (e-GFR), end-stage renal disease (ESRD) and/or death. Kaplan-Meier (K-M) analysis and multivariate Cox analysis were carried out to determine the predictors for renal outcomes. In order to eliminate the influence of different baseline data, a 1:2 propensity score (PS) match was performed to make the results comparable and convictive.
Results:
The baseline data suggested that patients in low serum bilirubin group had significantly higher levels of systolic blood pressure, proteinuria, serum creatinine and crescent formation ratio and lower levels of serum albumin and hemoglobin. Renal survival analysis indicated that lower serum bilirubin levels were significantly correlated with a poorer prognosis. Multivariate Cox analysis demonstrated that the higher level of serum bilirubin was an independent protective factor for renal survival (HR, 0.172; 95% CI, 0.030–0.991;
P
= 0.049). After PS matching, the baseline characters of two groups had no statistical differences. Similar outcomes were demonstrated in K-M curve and the multivariate Cox analysis.
Conclusion:
Elevated bilirubin levels might be related to the favorable renal outcomes.