Patients in the Intensive Care Unit (ICU) are usually wired to a bedside patient monitor known as a Multi-Parameter Patient Monitor (MPM). The MPM monitors one or more of the patient's vital physiological signals such as Electrocardiogram (ECG), Blood Oxygen Saturation (SpO2), Non-Invasive Blood Pressure (NiBP), Invasive Blood Pressures (IBP), Respiration, Temperature, Airway Gases, etc. Patient Monitoring Systems (PMS) are usually dedicated embedded systems, each with its own sensors, biomedical front-end hardware, LCD display, keypad, integrated thermal printer, communication interfaces such as RS232 or USB, and battery backup. These systems are quite expensive and thus contribute to the overall high costs of healthcare. These dedicated embedded systems have proprietary hardware and software. The biomedical Original Equipment Manufacturers (OEMs) avoid the use of general-purpose PC platforms for critical applications since the PC's hardware, software, and operating system (OS) are not optimized and reliable enough for medical applications. A typical MPM is shown in Figure 1.

Patient monitors have biomedical signal processing algorithms for detecting abnormal physiological conditions such as cardiac arrhythmias, apnea, low blood pressure, etc., and the monitor usually gives an audio-visual alert when such a condition is detected. The doctor usually intervenes when an alert is generated and either silences the system or takes corrective action to treat the patient. Apart from alerts generated by physiological conditions, other alerts are also generated to notify system status or malfunctions such as a low battery condition, sensor coming off the patient, etc. Since a doctor managing an ICU has to respond to alerts from several patients, it becomes an uphill task for the doctor to manage alerts, if the patient monitoring system does not have the capability to self-manage such alerts.

Our proposed framework addresses these issues by embedding intelligence at various levels in the system, in order that the system can manage itself and only require intervention by a doctor for critical physiological alerts.

As seen from Figure 1, the Computing and Results Display Block (CRDB) in a typical MPM is quite similar to that of a general-purpose PC. Hence, it makes sense to functionally separate the CRDB from the integrated PMS and use a single PC to cater to the computing and results display requirements of multiple patients.