| Despite the importance of controlling blood pressure (BP) to our health, BP remains inadequately managed worldwide. Due to global ageing and change of human lifestyles, the number of hypertensives is anticipated to continue rising from approximately 1 billion in 2000 to 1.56 billion by 2025. A stumbling block to BP management is high BP usually develops without obvious symptoms. As a result, many people are unaware of their condition until severe problems such as a stroke, a heart attack or kidney failure have occurred. In China, over 100 million people do not know they have developed hypertension and are living under a potential risk to their health. In addition to high BP, variations of BP are also independent indicators of morbidity and mortality of severe diseases. Yet, sudden changes in BP are difficult to be detected by state-of-the-art BP meters, which operate on principles that require an inflatable cuff to give only a snapshot of BP.;Consequently, this study aims to develop novel technologies that can measure BP non invasively and continuously without a cuff. The proposed method estimates BP using features including pulse transit time (PTT), which is defined as the time interval from R wave of electrocardiogram to onset of photoplethysmogram within the same heart cycle.;The proposed PTT-based technology was evaluated on 85 subjects (aged 57+/-29 yrs., including 39 hypertensives) whilst they were at rest in a sitting posture. A total of 999 pairs of systolic BP (SBP) and diastolic BP (DBP) estimations were made with reference to conventional cuff-based devices (i.e. a mercury sphygmomanometer and an oscillometric device) over a period of 6.4 weeks. The results of the study show that reference and estimated BP differed by 0.4+/-9.3 mmHg and 0.8+/-5.8 mmHg for SBP and DBP respectively (AAMI required mean and SD to be less than 5 and 8 mmHg correspondingly).;To further the investigation, a second study which was to investigate in a clinical setting for post-operation condition, was carried out on 8 patients (aged 55+/-18 yrs.) using the averaged invasive arterial-line and cuff readings taken at intervals of 40.0+/-24.7 min. as reference. After calibrating the new approach on each individual, it can estimate SBP and DBP within 3.3+/-6.5 mmHg and 4.3+/-6.4 mmHg of the reference for the complete set of 89 estimations.;The results of both studies show that the accuracy of the PTT-based technique is comparable to the cuff-based approaches. This technique is potentially useful to measure BP continuously.;Nevertheless, a major challenge of this approach is its requirement of a calibration procedure. One possible solution is to calibrate against a cuff-based device, but this is inconvenient particularly when calibration has to be refreshed from time to time. Therefore, a bio-model is proposed and developed for PTT along an artery where the hydrostatic component of BP varies. The model can be applied to calibrate the cuffless PTT-based approach and estimate BP by simple movements such as hand elevation. Several experiments were conducted to validate the assumptions of this model and the results were found to be promising.;Lastly, since the technology required information from several sensors that are placed on different body parts of a person, development of body area network (BAN) has been an important research focus. The concluding chapter of this thesis presents a new concept in this area, namely the hybrid body area network (h-BAN). In particular, the use of biological channels (bio-channels) for intra-BAN communication and securing wireless intra-BAN communication is discussed.;To conclude, this work developed a non-invasive and cuffless approach for BP measurement and addressed several key issues of this approach, i.e. the analysis, calibration, and implementation of it. The work can help to realise new BP management schemes in mobile health (m-Health) and personalised healthcare systems, which are developed to cater for the needs of the increasing aging population world-wide and to prevent and control chronic diseases like hypertension. |
