In 1817 James Parkinson gave the first English language description of the disease that now bears his name. Parkinson’s work gave a firm direction for others to follow. Sadly, despite some important advances, the causes of Parkinson’s disease remain unknown. There are no signs of a cure and existing therapies only mask the symptoms while the disease continues its grim, inexorable, progress.
Why is Parkinson’s disease so hard to solve?
A partial answer to this question lies in the fact that Parkinson’s is not a disease in the conventional sense – for example, no virus or bacteria are involved. Instead Parkinson’s seems to be associated with weaknesses in the internal processes that maintain the brain in goodorder. These weaknesses are caused by a number of things and can take many years before they present a critical risk. Advancing age is the most important factor that increases risk of Parkinson’s, but other risk factors exist. Traditional disease research is not suited to study conditions involving combinations of risk factors that can take a lifetime to develop. The time scales are too long and the number of combinatorial possibilities too great for experimental studies. In the case of Parkinson’s there is the added complication that direct experiments within the living human brain are impossible. And since only humans develop Parkinson’s disease, experiments with animal brains are of limited value.
Is there a better way of looking at Parkinson’s disease?
Multiple combinations of weaknesses and risks that can cause long-term failure are not unique to disease research. They occur everyday in engineering studies where systems (machines, structures and industrial processes) must be tested for reliability under all possible failure modes. Most of this testing is done before a system is built; using mathematical models it is possible to analyze and predict system performance under all possible operational circumstances. Using powerful computers it is possible to simulate and graphically visualise how an engineering system will behave as features of the system gradually change or fail. Simulating long term changes is not a problem, because computer models can simulate in fractions of a second combinations of events that might take decades in real-life.