Improving accuracy of vascular access quality classification in hemodialysis patients using deep learning with K highest score feature selection.

OBJECTIVE: To develop and evaluate a novel feature selection technique, using photoplethysmography (PPG) sensors, for enhancing the performance of deep learning models in classifying vascular access quality in hemodialysis patients. METHODS: This cross-sectional study involved creating a novel feature selection method based on SelectKBest principles, specifically designed to optimize deep learning models for PPG sensor data, in hemodialysis patients. The method effectiveness was assessed by comparing the performance of multiple deep learning models using the feature selection approach versus complete feature set. The model with the highest accuracy was then trained and tested using a 70:30 approach, respectively, with the full dataset and the SelectKBest dataset. Performance results were compared using Student's paired t-test. RESULTS: Data from 398 hemodialysis patients were included. The 1-dimensional convolutional neural network (CNN1D) displayed the highest accuracy among different models. Implementation of the SelectKBest-based feature selection technique resulted in a statistically significant improvement in the CNN1D model's performance, achieving an accuracy of 92.05% (with feature selection) versus 90.79% (with full feature set). CONCLUSION: These findings suggest that the newly developed feature selection approach might aid in accurately predicting vascular access quality in hemodialysis patients. This advancement may contribute to the development of reliable diagnostic tools for identifying vascular complications, such as stenosis, potentially improving patient outcomes and their quality of life.

DeepVAQ : an adaptive deep learning for prediction of vascular access quality in hemodialysis patients.

BACKGROUND: Chronic kidney disease is a prevalent global health issue, particularly in advanced stages requiring dialysis. Vascular access (VA) quality is crucial for the well-being of hemodialysis (HD) patients, ensuring optimal blood transfer through a dialyzer machine. The ultrasound dilution technique (UDT) is used as the gold standard for assessing VA quality; however, its limited availability due to high costs impedes its widespread adoption. We aimed to develop a novel deep learning model specifically designed to predict VA quality from Photoplethysmography (PPG) sensors. METHODS: Clinical data were retrospectively gathered from 398 HD patients, spanning from February 2021 to February 2022. The DeepVAQ model leverages a convolutional neural network (CNN) to process PPG sensor data, pinpointing specific frequencies and patterns that are indicative of VA quality. Meticulous training and fine-tuning were applied to ensure the model's accuracy and reliability. Validation of the DeepVAQ model was carried out against established diagnostic standards using key performance metrics, including accuracy, specificity, precision, F-score, and area under the receiver operating characteristic curve (AUC). RESULT: DeepVAQ demonstrated superior performance, achieving an accuracy of 0.9213 and a specificity of 0.9614. Its precision and F-score stood at 0.8762 and 0.8364, respectively, with an AUC of 0.8605. In contrast, traditional models like Decision Tree, Naive Bayes, and kNN demonstrated significantly lower performance across these metrics. This comparison underscores DeepVAQ's enhanced capability in accurately predicting VA quality compared to existing methodologies. CONCLUSION: Exemplifying the potential of artificial intelligence in healthcare, particularly in the realm of deep learning, DeepVAQ represents a significant advancement in non-invasive diagnostics. Its precise multi-class classification ability for VA quality in hemodialysis patients holds substantial promise for improving patient outcomes, potentially leading to a reduction in mortality rates.

COVID-19 incidence and outcomes among patients with kidney replacement therapy.

BACKGROUND: We aimed to investigate the incidence, fatality, and associated factors in patients with hemodialysis (HD), peritoneal dialysis (PD), and kidney transplantation (KT) hospitalized for coronavirus disease 2019 (COVID-19) infection and reimbursed from the National Health Security Office (NHSO). METHODS: The retrospective cohort analysis was conducted from an electronic-claimed database, and COVID-19 vaccination status was evaluated in patients with HD, PD, and KT from January 2020 to December 2021. There were 85,305 patients reimbursed for HD, PD, and KT by the NHSO. The rates of COVID-19 infection, COVID-19 vaccination, comorbidities, fatalities, and the cost of treatment were evaluated. RESULTS: COVID-19 infection was observed in 1,799 of 36,982 HD cases (4.9%), 1,531 of 45,453 PD cases (3.4%), and 95 of 2,870 KT cases (3.3%). Patients receiving COVID-19 vaccinations were most common in the KT group, followed by those with HD and PD (76.93% vs. 70.65% vs. 51.34%, respectively). KT patients had a lower fatality rate compared to those with PD and HD (8.42% vs. 18.41% vs. 21.40%, respectively). Advanced age, diabetes, cardiovascular diseases, and COVID-19 vaccination status were associated with fatality. The adjusted odds ratios of fatality after receiving one or two doses of vaccines were 0.7 (95% confidence interval [CI], 0.6-0.9) and 0.3 (95% CI, 0.2-0.4), respectively. The cost of treatment was highest in patients with HD, followed by PD and KT. CONCLUSION: The incidence of COVID-19 infection was higher in patients with HD than in those with PD or KT. COVID-19 vaccination following the national health policy should be encouraged for these patients to prevent fatality.

Time-dependent incidence rates and risk factors for transferring to hemodialysis in patients on peritoneal dialysis under the Thai PD-First Policy.

BACKGROUND: Haemodialysis (HD) transfer (HDT) is the major challenge of peritoneal dialysis (PD). This study aimed to analyse the time-dependent incidence rates and risk factors for permanent HDT in patients under Thailand's PD First policy. METHODS: The records of 20,545 patients from January 2008 to June 2018 were studied. The time on therapy (TOT) was divided into 0-3, 3-12, 12-24, 24-36, 36-48 and more than 48 months. The time-dependent incidence rates and causes of PD dropout were investigated. The risk factors for HDT were analysed by multivariable Poisson regression model and presented as incidence rate ratios (IRRs) and 95% confidence intervals (CIs). RESULTS: The main cause of PD dropout was death (45.7%) with 17.4% of the patients transferred to HD. The median (25th to 75th interquartile range) dialysis vintage was 1.4 (0.5-2.7) years. The incidence rates of HDT increased with TOT. Patients with universal coverage were transferred to HD less frequently than those with other health schemes. Patients who were illiterate or only had primary school education had a higher risk of being transferred to HD after 48 months of TOT (IRR 1.41 (95% CI 1.07-1.89)). Peritonitis within the first year of PD was the risk for HDT during 13-48 months of PD. The reasons for HDT changed with TOT. Mechanical complications followed by peritonitis were the main causes of HDT during the first 3 months, and after that peritonitis was the main reason. CONCLUSIONS: The incidence of HDT increased with TOT. The risks for HDT changed over time on PD.

The hidden financial catastrophe of chronic kidney disease under universal coverage and Thai "Peritoneal Dialysis First Policy".

Objective: Universal health coverage can decrease the magnitude of the individual patient's financial burden of chronic kidney disease (CKD), but the residual financial hardship from the patients' perspective has not been well-studied in low and middle-income countries (LMICs). This study aimed to evaluate the residual financial burden in patients with CKD stage 3 to dialysis in the "PD First Policy" under Universal Coverage Scheme (UCS) in Thailand. Methods: This multicenter nationwide cross-sectional study in Thailand enrolled 1,224 patients with pre-dialysis CKD, hemodialysis (HD), and peritoneal dialysis (PD) covered by UCS and other health schemes for employees and civil servants. We interviewed patients to estimate the proportion with catastrophic health expenditure (CHE) and medical impoverishment. The risk factors associated with CHE were analyzed by multivariable logistic regression. Results: Under UCS, the total out-of-pocket expenditure in HD was over two times higher than PD and nearly six times higher than CKD stages 3-4. HD suffered significantly more CHE and medical impoverishment than PD and pre-dialysis CKD [CHE: 8.5, 9.3, 19.5, 50.0% (p < 0.001) and medical impoverishment: 8.0, 3.1, 11.5, 31.6% (p < 0.001) for CKD Stages 3-4, Stage 5, PD, and HD, respectively]. In the poorest quintile of UCS, medical impoverishment was present in all HD and two-thirds of PD patients. Travel cost was the main driver of CHE in HD. In UCS, the adjusted risk of CHE increased in PD and HD (OR: 3.5 and 16.3, respectively) compared to CKD stage 3. Conclusions: Despite universal coverage, the residual financial burden remained high in patients with kidney failure. CHE was considerably lower in PD than HD, although the rates remained alarmingly high in the poor. The "PD First' program" could serve as a model for other LMICs. However, strategies to minimize financial distress should be further developed, especially for the poor.

Characteristics and Outcomes of Adults Hospitalized with Dengue Viral Infection and Acute Kidney Injury in Southern Thailand.

Dengue viral infection (DVI) among adult patients is increasingly problematic in tropical and subtropical regions. Acute kidney injury (AKI) after DVI poses substantial clinical outcomes and economic impact. This prospective study focused on the characteristics, risk factors, and outcomes of adult patients with AKI due to DVI hospitalized in nine network hospitals within Southern Thailand from January 2017 to December 2019. Among 120 adult patients hospitalized due to DVI without preexisting kidney diseases, 17 patients (14%) presented with AKI. During hospitalization, four patients required acute hemodialysis. The predominant characteristic of urinalysis was proteinuria, followed by pyuria and hematuria with remarkable dysmorphic red blood cells. Complications included acidosis, followed by hyperkalemia and volume overload. Most complications and deterioration of renal function occurred within the first week, but renal function recovered in second week of hospitalization. Stability of renal function was regained within the fourth week to the third month. However, four AKI patients recovered, with estimated glomerular filtration rate >60 mL/min/1.73 m2 within 3 months. Forty-day mortality rate and resource utilization, including hospital cost and length of hospitalization, among those with AKI were significantly higher than those without AKI. Thirty-day and in-hospital mortality rate among those with AKI was also higher than those without AKI. High APACHE II scores due to bleeding disorder and current use of non-steroidal anti-inflammatory agent were significantly associated with the emergence of AKI. Acute kidney injury among adult patients hospitalized due to DVI should be a concern and should be monitored for prompt treatment and follow-up.

Characteristics, Influencing Factors, Predictive Scoring System, and Outcomes of the Patients with Nephrotoxicity Associated with Administration of Intravenous Colistin.

Multidrug-resistant Gram-negative infection is a major global public health threat. Currently, colistin is considered the last-resort treatment despite its nephrotoxicity. The purpose of this study was to estimate the incidence, characteristics, and influencing factors and to develop a prediction model for colistin-associated nephrotoxicity. A retrospective study was conducted in the university hospital in the South of Thailand from December 2015 to June 2019. A total of 381 patients (median age (IQR) of 64 (51-62) years) were analyzed. Overall, 282 (74%) had nephrotoxicity according to the Kidney Disease: Improving Global Outcomes (KDIGO) classification. In-hospital, 30-day mortality rates and cost of hospital admission were significantly higher among those with nephrotoxicity. Age > 60 years, comorbidities, serum albumin less than 3.5 g/dL, and concomitant nephrotoxic use were significantly associated with colistin-associated nephrotoxicity with adjusted OR (95% CI) 2.01 (1.23-2.45), 1.85 (1.18-3.6), 1.68 (1.09-2.99), and 1.77 (1.10-2.97), respectively. The prediction model for high-risk colistin-associated nephrotoxicity was identified with good overall performance (specificity of 79.6% (95% CI 70.3-87.1) and positive predictive value of 92.1% (95% CI 88.0-95.1)). In conclusion, the incidence of colistin-associated nephrotoxicity was high and incurred significant morbidity, mortality, and economic burden. Our predictive scoring system is relatively simple and useful for optimizing colistin therapy.

National survey: Evaluation of cardiovascular risk factors in Thai patients with type 2 diabetes and chronic kidney disease after the development of cardiovascular disease.

AIM: The aim of the present study was to evaluate the achievement in controlling the risk factors of cardiovascular diseases (CVD) in Thai patients with type 2 diabetes (T2DM) and chronic kidney disease (CKD). METHODS: The DMHT dataset from 2011 to 2012, which was the cross-sectional study of the national survey in Thai patients with T2DM was analyzed. RESULTS: There were 1254 of 15 149 diabetic patients with estimated glomerular filtration rate (eGFR) less than 60 mL/min per 1.73 m2 that had developed CVD for more than 12 months. The prevalence of CVD was 8.3%. The mean age in years was 68.2 with a standard deviation (SD) of 8.7. Males and females were 38.7 and 61.3%, respectively. The mean duration of diabetes was 8.5 (SD 0.2) years. The mean body mass index was 25.5 (SD 4.4) kg/m2 . The percentage of patients with the target level of blood pressure control at ≤130/80 mm Hg was 47.1%. The percentage of patients who received angiotensin converting enzyme inhibitors (ACEI) or angiotensin receptor blockers (ARB) was 58.9%. The percentage of patients with the target level of LDL control at <70 mg/dL was 17.2%. The percentage of patients with the target level of HbA1C at 7% was 39.2%. There were 63 (5.0%) patients having recurrent CVD. CONCLUSIONS: Most Thai patients with T2DM and CKD with eGFR < 60 mL/min per 1.73 m2 could not achieve the therapeutic goals after the development of CVD. The national health policy should be planned to improve the quality of care to increase the number of patients who achieve the recommended goals.

A simplified clinical prediction score of chronic kidney disease: a cross-sectional-survey study.

BACKGROUND: Knowing the risk factors of CKD should be able to identify at risk populations. We thus aimed to develop and validate a simplified clinical prediction score capable of indicating those at risk. METHODS: A community-based cross-sectional survey study was conducted. Ten provinces and 20 districts were stratified-cluster randomly selected across four regions in Thailand and Bangkok. The outcome of interest was chronic kidney disease stage I to V versus non-CKD. Logistic regression was applied to assess the risk factors. Scoring was created using odds ratios of significant variables. The ROC curve analysis was used to calibrate the cut-off of the scores. Bootstrap was applied to internally validate the performance of this prediction score. RESULTS: Three-thousand, four-hundred and fifty-nine subjects were included to derive the prediction scores. Four (i.e., age, diabetes, hypertension, and history of kidney stones) were significantly associated with the CKD. Total scores ranged from 4 to 16 and the score discrimination was 77.0%. The scores of 4-5, 6-8, 9-11, and ≥ 12 correspond to low, intermediate-low, intermediate-high, and high probabilities of CKD with the likelihood ratio positive (LR+) of 1, 2.5 (95% CI: 2.2-2.7), 4.9 (95% CI: 3.9 - 6.3), and 7.5 (95% CI: 5.6 - 10.1), respectively. Internal validity was performed using 200 repetitions of a bootstrap technique. Calibration was assessed and the difference between observed and predicted values was 0.045. The concordance C statistic of the derivative and validated models were similar, i.e., 0.770 and 0.741. CONCLUSIONS: A simplified clinical prediction score for estimating risk of having CKD was created. The prediction score may be useful in identifying and classifying at riskpatients. However, further external validation is needed to confirm this.

Prevalence and risk factors of chronic kidney disease in the Thai adult population: Thai SEEK study.

BACKGROUND: Previous reports of chronic kidney disease (CKD) prevalence in Thailand varied from 4.3% to 13.8%. However, there were methodological concerns with these reports in terms of generalization and the accuracy of estimation. This study was, therefore, conducted to determine CKD prevalence and its risk factors in Thai adult populations. METHODS: The population-based Thai Screening and Early Evaluation of Kidney Disease (SEEK) study was conducted with cross-sectional stratified-cluster sampling. Serum creatinine was analysed using the modified Jaffe method and then standardized with isotope dilution mass spectrometry. RESULTS: The study included 3,459 subjects were included in the study. The mean age was 45.2 years (SE = 0.8), and 54.5% were female. Six hundred and twenty-six subjects were identified as having CKD, which evidenced an overall CKD prevalence of 17.5% [95% confidence interval (95% CI) = 14.6-20.4%]. The CKD prevalence of Stages I, II, III and IV were 3.3% (95% CI = 2.5%, 4.1%), 5.6% (95% CI = 4.2%, 7.0%), 7.5% (95% CI = 6.2%, 8.8%) and 1.1% (95% CI = 0.7%, 1.5%), respectively. The prevalence of CKD was higher in Bangkok, the Northern and Northeastern regions than in the Central and Southern regions. Seven factors (i.e. age, gender, diabetes, hypertension, hyperuricaemia, history of kidney stones and the use of traditional medicines) were associated with CKD. Only 1.9% of the subjects were aware that they had CKD. CONCLUSIONS: CKD prevalence in the Thai population is much higher than previously known and published. Early stages of CKD seem to be as common as later stages. However, albuminuria measurement was not confirmed and adjusting for persistent positive rates resulted in the prevalence of 14.4%. Furthermore, the awareness of CKD was quite low in the Thai population.

Peritonitis in acute peritoneal dialysis in a university hospital.

OBJECTIVE: Although acute peritoneal dialysis is a useful procedure, peritonitis is often a complication. When the patient is mainly at risk of peritonitis is controversial. The purpose of this study was to find the incidence time of peritonitis, the infecting microorganism, and risk factors. DESIGN: A retrospective study. PATIENTS: 118 cases of acute peritoneal dialysis in 93 patients were included in this study. METHOD: Data were collected from medical records. RESULTS: Overall, the peritonitis rate was 36.45 per cent. The peritonitis rate rose following the duration of dialysis from 11 per cent on the first day to 21 per cent on the third day, although the difference was not statistically significant. Gram-negative bacilli were predominant, at 81.6 per cent. Acinetobacter baumanii and Enterobacter cloacae were the two most common organisms (23.7 and 21.1% respectively). There was a significantly higher male to female ratio in the peritonitis group than the no-peritonitis group (3.33:1 and 1.2:1 respectively, p=0.028). CONCLUSION: There was a high peritonitis rate in acute peritoneal dialysis. The most common microorganisms were gram-negative bacilli, Acinetobacter baumanii and Enterobacter cloacae. The risk factor was male sex. Duration of dialysis of more than 2 days tended to increase the risk of peritonitis.

Increased incidence of benign breast disease in female renal transplant patients receiving cyclosporin.

BACKGROUND: Unlike other cancers, breast cancer does not occur at increased frequency in renal transplant patients but fibroadenomata may be more common as a result of exposure to cyclosporin. In order to determine the incidence of benign breast disease in renal transplant patients at Monash Medical Centre, current female patients were studied. METHODS: The study was divided into two parts: (i) a retrospective review of those who presented with clinically detectable breast lumps; and (ii) mammographic screening of current female transplant patients who had been transplanted for more than 1 year. RESULTS: In the retrospective study there were 11 patients with 16 breast lumps among a total of 85 patients. All were confirmed by biopsy. The mean age at diagnosis of breast lumps was 41.5 years (range 25-70 years). The mean time to presentation was 3.5 years after transplantation. Nine out of 11 patients had benign breast disease including fibroadenoma (six patients), fibrocystic disease (two patients) and intraductal papillomatosis (one patient). Two patients had breast cancer. Five of the patients with fibroadenoma had multiple lumps and a recurrent course. All patients with fibroadenomata had received cyclosporin. In the second part, 54 patients were further screened. The mean duration of transplantation was 6.4 years (range 1.25-18.5 years). Eighty-seven per cent of the patients had received cyclosporin, and 80% had a negative (normal) study. Seven of 54 had abnormalities including cysts and calcification, of whom two patients had fibroadenomata. Four patients had 'dense mammograms', all of whom received cyclosporin as a part of their immunosuppression. No breast cancer was detected during the study. CONCLUSION: The incidence of benign breast disease in the female transplant patients studied was far greater then the general population. The increase in fibroadenomata, in particular, may relate to the use of cyclosporin.