The role of real-time in biomedical science: A meta-analysis on computational complexity, delay and speedup. (1st March 2015)

• Record Type: Journal Article
• Authors: Faust, Oliver; Yu, Wenwei; Rajendra Acharya, U.
• Is Part Of: Computers in biology and medicine. Volume 58(2015)
• Page Start: 73
• Page End: 84
• Publication Date: 2015-03-01
• Subjects: Real-time -- Computational complexity -- Speedup -- Delay -- Health care systems; A algorithm -- AA application area -- CC computational complexity -- CT computed tomography -- CPU central processing unit -- Δ delay -- DNA deoxyribonucleic acid -- EMB embedded -- GA genetic analysis -- GPU graphics processing unit -- H hardware complexity -- HRV heart rate variability -- I implementation -- I/A implementation/algorithm -- Im image processing -- O big O notation -- PCR polymerase chain reaction -- PT processing time -- RAM random-access memory -- Si signal processing -- SIM simulation environment -- Sp speedup -- VLSI very large-scale integration; Medicine -- Data processing -- Periodicals; Biology -- Data processing -- Periodicals; 610.285
• DOI: 10.1016/j.compbiomed.2014.12.024
• Languages: English
• ISSNs: 0010-4825

Abstract

Abstract: Purpose: The concept of real-time is very important, as it deals with the realizability of computer based health care systems. Method: In this paper we review biomedical real-time systems with a meta-analysis on computational complexity (CC), delay ( Δ ) and speedup ( S p ). Results: During the review we found that, in the majority of papers, the term real-time is part of the thesis indicating that a proposed system or algorithm is practical. However, these papers were not considered for detailed scrutiny. Our detailed analysis focused on papers which support their claim of achieving real-time, with a discussion on CC or S p . These papers were analyzed in terms of processing system used, application area (AA), CC, Δ, S p, implementation/algorithm (I/A) and competition. Conclusions: The results show that the ideas of parallel processing and algorithm delay were only recently introduced and journal papers focus more on Algorithm (A) development than on implementation (I). Most authors compete on big O notation ( O ) and processing time ( PT ). Based on these results, we adopt the position that the concept of real-time will continue to play an important role in biomedical systems design. We predict that parallel processing considerations, such as S p and algorithm scaling, will become more important. Abstract : Highlights: Real-time: The promise of performance in real world scenarios. In this review we have discussed 120 real-time processing papers in the biomedical engineering field. The majority of papers lacks appropriate measures to support the real-time claim. An appropriate measure for theoretical claims on real-time is computational complexity. An appropriate measure for practical claims on real-time is speedup.