Computer Aided Detection Method for Ischemic Stroke Using Feature Based Approach

The human brain is the most complex and metabollically demanding organ of the entire body. It is a jelly-like mass of tissue comprising about 2.5% of the total body weight [1]. It takes about 20% of the body’s oxygen consumption. The brain consists of forebrain, midbrain and hindbrain. It controls the central nervous system (CNS) via the cranial nerves and spinal cord. It also controls the peripheral nervous system (PNS) and regulates virtually all human activities. We can control our mental activity consciously such as thought, action, memory, feeling and experience of the world. A disturbance to the cerebral blood vessels or cerebral blood flow will lead to disastrous consequences. The vascular disease of the brain is called cerebrovascular accident (CVA) or stroke.

With the aging of our population, stroke has been rated the third highest cause of death (second to heart disease and cancer) in the world. In the USA, it was estimated that there were about 700,000 new and recurrent cases of stroke per year [2]. In Hong Kong, there were around 16,000 new/recurrent cases of stroke and more than 3,000 people died due to stroke per year [3]. Stroke results from insufficient supply of oxygen and nutrient to the brain tissues. The nerve cells die, and there is no replacement of nerve cells. Therefore, the nervous system will deteriorate and this will affect the normal functions of the body. As a result, the loss of function appears rapidly even within minutes. The major types of stroke are ischemic stroke and hemorrhagic stroke. The symptoms of stroke include acute hemi-paresis, aphasia, non-focal neurological deflect, etc. Most cases involve ischemic stroke rather than hemorrhagic stroke in the ratio of 80% versus 20%. In addition, ischemic stroke is more common in the Chinese community (Ho, 2002). Among these patients, only about 10% could fully recover. Therefore, this has been a leading cause of severe and long-term disability with huge direct and indirect cost to society [2].

Diagnosis of the type of stroke must be accurate. In the emergency department, 20% of the diagnosis might be incorrect [4]. In addition to the neurological examination, if the Glasgow Coma Scale (GCS) [5] or National Institutes of Health Stroke Scale (NIHSS) [6] lies somewhere between severe and moderate level, patients are required to undergo an emergency brain scan, for example, Computed Tomography (CT) or Magnetic Resonance Imaging (MRI). Other imaging modalities such as Positron Emission Tomography (PET) and Single Photon Emission Computed Tomography (SPECT) are available for diagnosis of brain diseases.