There is a rise of using mathematical models to treat patients.  It is hoped that by having a computational model of biological processes, it can become easier to diagnose and treat patients. 

 One example is the HIV virus.  This virus, transcripted by a retrovirus, is known to evolve rapidly.  Thus a drug that targets a certain variant becomes ineffective due to the rapid mutation.  A certain method is using the patient’s genotype and phenotype to determine the variants present inside the patient.  Simulations are run to determine the mutation rates and the evolution of variants.  This can aid in determining the best course of treatment for the patient. 

Another example is used in predicting and diagnosing heart disease.  Most tests are used to determine activity at the surface of the heart.  By taking into account the geometry of the art and information from surface tests, models can be developed to simulate and study the human heart in patients.  This can aid in diagnosis and treatment of diseases.   The equations of the processes that evolve in the heart are already present.  By integrating these, the computational models can be developed and refined.  Perhaps, virtual proposed therapies can be applied and tested on a virtual heart.  This can determine a better course of action in the actual human patient. 

 Another area is mapping the genome and modeling which genes may be more responsible for certain diseases.  In this way in the future, human predispositions to certain illnesses can be determined and a corrective course of preventive action and treatment can be undertaken . 

 http://www.scientific-computing.com/news/news_story.php?news_id=333

http://www.scientific-computing.com/news/news_story.php?news_id=290

http://www.scientific-computing.com/news/news_story.php?news_id=313