A web-based clinical tool for predicting occult bleeding in dengue patients based on haematocrit changes and fluid balance.
🔗 Live Calculator: https://SAIFULSAFUAN.COM/bleeders/
The Dengue Bleeder Calculator implements a mathematical model to help healthcare professionals identify potential occult bleeding in dengue patients. By comparing observed haematocrit values with estimated values based on fluid balance, the tool can flag cases where undetected bleeding may be occurring.
In dengue fever, two major complications affect haematocrit levels:
This calculator helps differentiate between these conditions by accounting for fluid balance and comparing expected vs. observed haematocrit values.
The calculator uses the following core formula:
Hct₁ = (Hct₀ × Weight × F) / (Weight × F + Balance)
Where:
Hct₁ = Estimated haematocrit after fluid shiftHct₀ = Baseline haematocritWeight = Patient weight (kg)F = Correction factor (75 ml/kg for males, 65 ml/kg for females)Balance = Fluid input - Fluid output (ml)The tool calculates a lower limit threshold:
Lower Limit = Hct₁ - (Hct₀ - Hct₁) × x
Where x is an adjustable correction factor for insensible losses (0-0.5).
Interpretation:
⚠️ Important Notes:
This calculator is based on the mathematical model described in:
“Underlying Mathematics to the Dengue Bleeder Calculator/Predictor Tool”
The model leverages the principle that in dengue:
[GNU Affero 3.0]
This calculator is intended for use by qualified healthcare professionals as a clinical decision support tool. It should not be used as the sole basis for diagnosis or treatment decisions. Always correlate with clinical findings and other laboratory results.
Quick Start: Visit https://SAIFULSAFUAN.COM/bleeders/ to use the calculator immediately.
Developed with the aim of improving dengue patient care through early detection of occult bleeding.
🔗 Live Calculator: https://SAIFULSAFUAN.COM/denguedeathpredictor/
This web application implements a clinical decision support tool for predicting mortality risk in dengue patients using the REPROSED2017 Elastic Net Model. The calculator is based on a glmnet (elastic net regularized logistic regression) model trained on clinical parameters commonly available in dengue patient care settings.
The model uses 8 key clinical parameters:
| Parameter | Description | Units | Clinical Significance |
|---|---|---|---|
| Persistent Diarrhea | Presence of ongoing diarrhea | Yes/No | Indicator of dengue with GI involvement |
| BMI | Body Mass Index | kg/m² | Nutritional status and body composition |
| RR | Respiratory Rate | breaths/min | Indicator of respiratory distress or compensation |
| Platelets | Platelet count | ×10³/µL | Marker of dengue severity and bleeding risk |
| HCO₃ | Serum bicarbonate | mmol/L | Acid-base status, metabolic acidosis indicator |
| Lactate | Serum lactate | mmol/L | Tissue perfusion and shock severity |
| Albumin | Serum albumin | g/L | Capillary leak and nutritional status |
| AST | Aspartate aminotransferase | U/L | Liver involvement and organ dysfunction |
The model was developed as part of the REPROSED2017 study, which aimed to create a reliable prediction tool for dengue mortality in severe dengue cases using readily available clinical parameters.
The elastic net approach helps to:
Important Note: This demonstration uses placeholder coefficients. For clinical use, the actual model coefficients must be extracted from the glmnetfin31.rda file using:
# In R, load the model and extract coefficients
load("glmnetfin31.rda")
coefficients <- coef(glmnetfin31$finalModel, s = glmnetfin31$bestTune$lambda)
print(coefficients)
This calculator should be used as a decision support tool in conjunction with clinical judgment. It is designed for:
[GNU Affero 3.0]
Clinical decisions should not be based solely on this calculator’s output. Always use clinical judgment and consider the full clinical context when managing dengue patients.
For questions, suggestions, or bug reports, please: