Treatment protocol for HIV patients is dependent upon their HIV viral load (VL) which tends to fluctuate from 0 (i.e. undetectable) to 107 copies per ml over repeated measurements. VL also plays a major role in determining efficiency of anti-retroviral therapy (ART) and it is hence important to develop an objective measure of VL level and classifying patients based on time-varying patterns of VL. Farooq et. al. used relative area of viral exposure, weighted recency reliability (wRR), adjusted maximal difference, and interquartile range (IQR) in order to capture the time-varying VL patterns. A unique centroid algorithm has been employed to categorize HIV positive patients depending upon such patterns into one of the five categories: sustained low viral load (SLVL), durably suppressed viral load (DSVL), high viral load suppression (HVLS), rebounding viral load (RVL) and sustained high viral load (SHVL). This is a novel approach for patient segmentation and the same method can be used for modelling other viral infections.
Input variables : Viral load data
Output Variables : types of viral load patterns: durably suppressed viral load (DSVL), sustained low viral load (SLVL), sustained high viral load (SHVL), high viral load suppression (HVLS), rebounding viral load (RVL)
Visit Model : arxiv.org
Additional links : arxiv.org
| Model Category | : | Public |
| Date Published | : | April, 2018 |
| Healthcare Domain | : |
Medical Technology
Provider |
| Code | : | github.com |
| Health Risk Management |
| Disease Detection |