From: Application of deep learning to predict the low serum albumin in new hemodialysis patients
Characteristics | Comparison | Fully adjusted | GOA feature selection | ||||
---|---|---|---|---|---|---|---|
OR | 95% CI | P | OR | 95% CI | P | ||
Age | Years | 1.03 | 1.01–1.04 |  < 0.001 | 1.03 | 1.02–1.04 | < 0.001 |
Gender | Male v.s Female | 1.41 | 1.00–1.99 | 0.051 | 1.48 | 1.07–2.06 | 0.018 |
Comorbidity | Â | Â | Â | Â | Â | Â | Â |
 Diabetes mellitus | Yes v.s No | 0.85 | 0.59–1.2 | 0.356 |  |  |  |
 Hypertension | Yes v.s No | 1.05 | 0.70–1.60 | 0.805 | 1.01 | 0.68–1.52 | 0.965 |
 Heart Failure | Yes v.s No | 1.05 | 0.74–1.49 | 0.778 |  |  |  |
 Cancer | Yes v.s No | 1.14 | 0.84–1.56 | 0.389 |  |  |  |
Laboratory measurements | Â | Â | Â | Â | Â | Â | Â |
 Hb | g/dL | 0.85 | 0.74–0.96 | 0.012 | 0.83 | 0.73–0.95 | 0.006 |
 Fe | ug/dL | 0.99 | 0.98–0.99 |  < 0.001 | 0.99 | 0.98–0.99 | < 0.001 |
 Ferritin | ug/dL | 1.001 | 1.0003–1.0010 |  < 0.001 | 1.001 | 1.0004–1.0011 | < 0.001 |
 Na | mEq/L | 0.94 | 0.90–0.98 | 0.005 | 0.94 | 0.90–0.98 | 0.005 |
 K | mEq/L | 0.79 | 0.62–0.99 | 0.045 | 0.79 | 0.64–0.98 | 0.037 |
 Ca | mg/dL | 0.71 | 0.59–0.86 | 0.001 | 0.72 | 0.59–0.86 | 0.001 |
 Phosphate | mg/dL | 0.99 | 0.87–1.13 | 0.873 |  |  |  |
 BUN | mg/dL | 1.00 | 0.99–1.01 | 0.951 |  |  |  |
 Cr | mg/dL | 0.77 | 0.71–0.84 |  < 0.001 | 0.77 | 0.71–0.83 | < 0.001 |
 Alkaline phosphatase | U/L | 1.01 | 1.00–1.01 |  < 0.001 | 1.01 | 1.00–1.01 | < 0.001 |
 iPTH | pg/dL | 1.00 | 0.9994–1.0005 | 0.951 |  |  |  |
 Cholesterol | mg/dL | 0.997 | 0.99–1.00 | 0.259 |  |  |  |
 Triglyceride | mg/dL | 0.999 | 0.9971–1.0003 | 0.133 | 0.998 | 0.9968–0.9998 | 0.030 |
 Fasting glucose (AC) | mg/dL | 1.00 | 0.9981–1.0022 | 0.885 |  |  |  |
Optimal AIC | Â | 1173.52 | 1160.71 |