Stream-based Biomedical Classification Algorithms for Analyzing Biosignals

Fong, Simon,Hang, Yang,Mohammed, Sabah,Fiaidhi, Jinan Korea Information Processing Society 2011 Journal of information processing systems Vol.7 No.4

Classification in biomedical applications is an important task that predicts or classifies an outcome based on a given set of input variables such as diagnostic tests or the symptoms of a patient. Traditionally the classification algorithms would have to digest a stationary set of historical data in order to train up a decision-tree model and the learned model could then be used for testing new samples. However, a new breed of classification called stream-based classification can handle continuous data streams, which are ever evolving, unbound, and unstructured, for instance--biosignal live feeds. These emerging algorithms can potentially be used for real-time classification over biosignal data streams like EEG and ECG, etc. This paper presents a pioneer effort that studies the feasibility of classification algorithms for analyzing biosignals in the forms of infinite data streams. First, a performance comparison is made between traditional and stream-based classification. The results show that accuracy declines intermittently for traditional classification due to the requirement of model re-learning as new data arrives. Second, we show by a simulation that biosignal data streams can be processed with a satisfactory level of performance in terms of accuracy, memory requirement, and speed, by using a collection of stream-mining algorithms called Optimized Very Fast Decision Trees. The algorithms can effectively serve as a corner-stone technology for real-time classification in future biomedical applications.

Influence of geology and human activity on the genetic structure and demography of the Oriental fire-bellied toad (Bombina orientalis)

Fong, J.J.,Li, P.P.,Yang, B.T.,Zhou, Z.Y.,Leache, A.D.,Min, M.S.,Waldman, B. Academic Press 2016 Molecular phylogenetics and evolution Vol.97 No.-

The Oriental fire-bellied toad (Bombina orientalis) is a commonly used study organism, but knowledge of its evolutionary history is incomplete. We analyze sequence data from four genetic markers (mtDNA genes encoding cytochrome c oxidase subunit I, cytochrome b, and 12S-16S rRNA; nuDNA gene encoding recombination activating gene 2) from 188 individuals across its range in Northeast Asia to elucidate phylogeographic patterns and to identify the historic events that shaped its evolutionary history. Although morphologically similar across its range, B. orientalis exhibits phylogeographic structure, which we infer was shaped by geologic, climatic, and anthropogenic events. Phylogenetic and divergence-dating analyses recover four genetically distinct groups of B. orientalis: Lineage 1-Shandong Province and Beijing (China); Lineage 2-Bukhan Mountain (Korea); Lineage 3-Russia, Northeast China, and northern South Korea; and Lineage 4-South Korea. Lineage 2 was previously unknown. Additionally, we discover an area of secondary contact on the Korean Peninsula, and infer a single dispersal event as the origin of the insular Jeju population. Skyline plots estimate different population histories for the four lineages: Lineages 1 and 2 experienced population decreases, Lineage 3 remained stable, while Lineage 4 experienced a sharp increase during the Holocene. The timing of the population expansion of Lineage 4 coincides with the advent of rice cultivation, which may have facilitated the increase in population size by providing additional breeding habitat.

Wheelchair martial arts practitioners have similar bone strength, sitting balance and self-esteem to healthy individuals

Fong, Shirley S.M.,Ng, Shamay S.M.,Li, Anthony O.T.,Guo, X. korean Academy of Physical Therapy Rehabilitation 2014 Physical therapy rehabilitation science Vol.3 No.1

Objective: The aim of this study was to compare the radial bone strength, sitting balance ability and global self-esteem of wheelchair martial arts practitioners and healthy control participants. Design: Cross-sectional study. Methods: Nine wheelchair martial art practitioners with physical disabilities and 28 able-bodied healthy individuals participated in the study. The bone strength of the distal radius was assessed using the Sunlight Mini-Omni Ultrasound Bone Sonometer; sitting balance was quantified using the modified functional reach test (with reference to a scale marked on the wall); and the self-administered Rosenberg self-esteem (RSE) scale was used to measure the global self-esteem of the participants. The velocity of the ultrasound wave (speed of sound, m/s) traveling through the outer surface of the radial bone was measured and was then converted into a T-score and a Z-score. These ultrasound T-score and Z-score that represent bone strength; the maximum forward reaching distance in sitting (cm) that represents sitting balance; and the RSE total self-esteem score that indicates global self-esteem were used for analysis. Results: The results revealed that there were no statistically significant between-group differences for radial bone-strength, maximum forward reaching distance, or self-esteem outcomes. Conclusions: The wheelchair martial arts practitioners had similar radial bone strength, sitting balance performance and self-esteem to able-bodied healthy persons. Our results imply that wheelchair martial arts might improve bone strength, postural control and self-esteem in adult wheelchair users. This new sport-wheelchair martial arts-might be an exercise option for people with physical disabilities.

Cervical Screening Using Visual Inspection with Acetic Acid (VIA) and Treatment with Cryotherapy in Fiji

Fong, James,Gyaneshwar, Rajaneshwar,Lin, Sophia,Morrell, Stephen,Taylor, Richard,Brassil, Ann,Stuart, Anne,McGowan, Catherine Asian Pacific Journal of Cancer Prevention 2014 Asian Pacific journal of cancer prevention Vol.15 No.24

The purpose of this study was to demonstrate the feasibility of VIA screening with cryotherapy and to record normative values for indicators anticipated in similar low resource settings. Women aged 30-49 years were targeted, resulting in 1961 women screened and treated at two primary health care (PHC) centres near Suva, Fiji. Recruitment was through provision of information, education and communication (IEC). Referrals to a gynaecology outpatient department (OPD) at a referral hospital occurred throughout the screening pathway. Participation was 32% (95%CI 31-33%), higher in iTaukei (Melanesians) women (34%, 95%CI 33-36) compared to Fijians of Indian descent (26%, 95%CI 24-28). Regression analysis, adjusted for confounders, indicated significantly lower participation in those of Indian descent, and age groups 35-39 and 45-49 years. Of those examined by VIA, 190 were positive with aceto-white lesions (9.9%), within the expected range of 8-15%, with minor geographic and ethnic variation. Positive VIA results were more common in the peri-urban area, and in those aged 35-39 years. Of women aged 30-49 years, 59 received cryotherapy (none of whom had significant complications), 91 were referred to OPD, two cervical carcinomas were identified and eight cervical intra-epithelial neoplasms (CIN) II-III were diagnosed. These results provide normative findings from a community-based VIA screening program for other similar low resource settings.

A time series pre-processing methodology with statistical and spectral analysis for classifying non-stationary stochastic biosignals

Fong, S.,Cho, K.,Mohammed, O.,Fiaidhi, J.,Mohammed, S. Springer Science + Business Media 2016 The Journal of supercomputing Vol.72 No.10

<P>Biosignal classification is an important non-invasive diagnosis tool in biomedical application, e.g. electrocardiogram (ECG), electroencephalogram (EEG), and electromyogram (EMG) that helps medical experts to automatically classify whether a sample of biosignal under test/monitor belongs to the normal type or otherwise. Most biosignals are stochastic and non-stationary in nature, that means their values are time dependent and their statistics vary over different points of time. However, most classification algorithms in data mining are designed to work with data that possess multiple attributes to capture the non-linear relationships between the values of the attributes to the predicted target class. Therefore, it has been a crucial research topic for transforming univariate time series to multivariate dataset to fit into classification algorithms. For this, we propose a pre-processing methodology called statistical feature extraction (SFX). Using the SFX we can faithfully remodel statistical characteristics of the time series via a sequence of piecewise transform functions. The new methodology is tested through simulation experiments over three representative types of biosignals, namely EEG, ECG and EMG. The experiments yield encouraging results supporting the fact that SFX indeed produces better performance in biosignal classification than traditional analysis techniques like Wavelets and LPC-CC.</P>

Effects of Chlorella on Activities of Protein Tyrosine Phosphatases, Matrix Metalloproteinases, Caspases, Cytokine Release, B and T Cell Proliferations, and Phorbol Ester Receptor Binding

Fong-Chi Cheng,Atsui Lin,Jin-Jye Feng,Toru Mizoguchi,Hideo Takekoshi,Hitoshi Kubota,Yoko Kato,Yo Naoki 한국식품영양과학회 2004 Journal of medicinal food Vol.7 No.2

Improvised methods for tackling big data stream mining challenges: case study of human activity recognition

Fong, S.,Liu, K.,Cho, K.,Wong, R.,Mohammed, S.,Fiaidhi, J. Springer Science + Business Media 2016 The Journal of supercomputing Vol.72 No.10

<P>Big data stream is a new hype but a practical computational challenge founded on data streams that are prevalent in applications nowadays. It is quite well known that data streams that are originated and collected from monitoring sensors accumulate continuously to a very huge amount making traditional batch-based model induction algorithms infeasible for real-time data mining or just-in-time data analytics. In this position paper, following a new data stream mining methodology, namely stream-based holistic analytics and reasoning in parallel (SHARP), a list of data analytic challenges as well as improvised methods are looked into. In particular, two types of decision tree algorithms, batch-mode and incremental-mode, are put under test at sensor data that represents a typical big data stream. We investigate whether and to what extent of two improvised methods-outlier removal and balancing imbalanced class distributions-affect the prediction performance in big data stream mining. SHARP is founded on incremental learning which does not require all the training to be loaded into the memory. This important fundamental concept needs to be supported not only by the decision tree algorithms, but by the other improvised methods usually at the preprocessing stage as well. This paper sheds some light into this area which is often overlooked by data analysts when it comes to big data stream mining.</P>

Great Power Competition in Non‐sovereign Entities: US–China Tug‐of‐War over Hong Kong, 1950–2020

Fong Brian C. H. 인하대학교 국제관계연구소 2022 Pacific Focus Vol.37 No.2

The theory of offensive realism postulates that when a potential hegemon and an existing hegemon clash, their intense security competition will span all corners of the globe. This study advances the literature by investigating the impact of great power on non-sovereign entities, using Hong Kong as a case study. Drawing on archival and published data, this study examines the geopolitical status of Hong Kong as a function of the changing US–China great power relations. The findings of this study demonstrate that as a consequence of the changing US–China relations, Hong Kong has been transformed from a geopolitical buffer zone in the 1950s and 1960s (US–China restrained competition) and the 1970s through the 2000s (US–China rapprochement) into a geopolitical hot spot since the 2010s (US–China intense competition). This study contributes to the literature by fostering dialogues between great power studies and the non-sovereignty literature.

Avoiding complications in microsurgery and strategies for flap take-back

Fong, Hui Chai,Levin, Lawrence Scott Korean Society of Plastic and Reconstructive Surge 2019 Archives of Plastic Surgery Vol.46 No.5

A Comparison between Structural Allografts and Polyetheretherketone Interbody Spacers Used in Anterior Cervical Discectomy and Fusion: A Systematic Review and Meta-analysis

Fong Francis Jia Yi,Lim Chee Yit,Tan Jun-Hao,Hey Hwee Weng Dennis 대한척추외과학회 2024 Asian Spine Journal Vol.18 No.1

Among interbody implants used during anterior cervical discectomy and fusion (ACDF), structural allografts and polyetheretherketone (PEEK) are the most used spacers. Currently, no consensus has been established regarding the superiority of either implant, with US surgeons preferring structural allografts, whereas UK surgeons preferring PEEK. The purpose of this systematic review (level of evidence, 4) was to compare postoperative and patient-reported outcomes between the use of structural allografts PEEK interbody spacers during ACDF. Five electronic databases (PubMed, Embase, Scopus, Web of Science, and Cochrane) were searched for articles comparing the usage of structural allograft and PEEK interbody spacers during ACDF procedures from inception to April 10, 2023. The searches were conducted using the keywords “Spine,” “Allograft,” and “PEEK” and were performed in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-analysis guidelines. Subsequent quality and sensitivity analyses were performed on the included studies. Nine studies involving 1,074 patients were included. Compared with the PEEK group, the structural allograft group had comparable rates of postoperative pseudoarthrosis (<i>p</i>=0.58). However, when stratified according to the number of levels treated, the 3-level ACDF PEEK group was 3.45 times more likely to have postoperative pseudoarthrosis than the structural allograft group (<i>p</i>=0.01). Subsequent postoperative outcomes (rate of subsidence and change in the preoperative and postoperative segmental disc heights) were comparable between the PEEK and structural allograft groups. Patient-reported outcomes (Visual Analog Scale [VAS] of neck pain and Neck Disability Index [NDI]) were comparable. This study showed that for 3-level ACDFs, the use of structural allografts may confer higher fusion rates. However, VAS neck pain, NDI, and subsidence rates were comparable between structural allografts and PEEK cages. In addition, no significant difference in pseudoarthrosis rates was found between PEEK cages and structural allografts in patients undergoing 1- and 2-level ACDFs.