liu.seSearch for publications in DiVA
Change search
Link to record
Permanent link

Direct link
Pham, Tuan, ProfessorORCID iD iconorcid.org/0000-0002-4255-5130
Alternative names
Biography [eng]

I am a Professor of Biomedical Engineering.  My current research focuses on image processing, time-series analysis and pattern recognition applied to medicine, biology, and mental health.  

Publications (10 of 114) Show all publications
Pham, T., Wårdell, K., Eklund, A. & Salerud, G. (2019). Classification of Short Time Series in Early Parkinson’s Disease With Deep Learning of Fuzzy Recurrence Plots. IEEE/CAA Journal of Automatica Sinica, 6(6), 1306-1317
Open this publication in new window or tab >>Classification of Short Time Series in Early Parkinson’s Disease With Deep Learning of Fuzzy Recurrence Plots
2019 (English)In: IEEE/CAA Journal of Automatica Sinica, ISSN 2329-9266, Vol. 6, no 6, p. 1306-1317Article in journal (Refereed) Published
Abstract [en]

There are many techniques using sensors and wearable devices for detecting and monitoring patients with Parkinson's disease (PD). A recent development is the utilization of human interaction with computer keyboards for analyzing and identifying motor signs in the early stages of the disease. Current designs for classification of time series of computer-key hold durations recorded from healthy control and PD subjects require the time series of length to be considerably long. With an attempt to avoid discomfort to participants in performing long physical tasks for data recording, this paper introduces the use of fuzzy recurrence plots of very short time series as input data for the machine training and classification with long short-term memory (LSTM) neural networks. Being an original approach that is able to both significantly increase the feature dimensions and provides the property of deterministic dynamical systems of very short time series for information processing carried out by an LSTM layer architecture, fuzzy recurrence plots provide promising results and outperform the direct input of the time series for the classification of healthy control and early PD subjects.

Keywords
Deep learning, early Parkinson’s disease (PD), fuzzy recurrence plots, long short-term memory (LSTM) neural networks, pattern classification, short time series
National Category
Medical Engineering
Identifiers
urn:nbn:se:liu:diva-161818 (URN)10.1109/JAS.2019.1911774 (DOI)000503189200003 ()
Available from: 2019-11-11 Created: 2019-11-11 Last updated: 2020-06-11Bibliographically approved
Jin, Q., Meng, Z., Pham, T., Chen, Q., Wei, L. & Su, R. (2019). DUNet: A deformable network for retinal vessel segmentation. Knowledge-Based Systems, 178, 149-162
Open this publication in new window or tab >>DUNet: A deformable network for retinal vessel segmentation
Show others...
2019 (English)In: Knowledge-Based Systems, ISSN 0950-7051, E-ISSN 1872-7409, Vol. 178, p. 149-162Article in journal (Refereed) Published
Abstract [en]

Automatic segmentation of retinal vessels in fundus images plays an important role in the diagnosis of some diseases such as diabetes and hypertension. In this paper, we propose Deformable U-Net (DUNet), which exploits the retinal vessels’ local features with a U-shape architecture, in an end to end manner for retinal vessel segmentation. Inspired by the recently introduced deformable convolutional networks, we integrate the deformable convolution into the proposed network. The DUNet, with upsampling operators to increase the output resolution, is designed to extract context information and enable precise localization by combining low-level features with high-level ones. Furthermore, DUNet captures the retinal vessels at various shapes and scales by adaptively adjusting the receptive fields according to vessels’ scales and shapes. Public datasets: DRIVE, STARE, CHASE_DB1 and HRF are used to test our models. Detailed comparisons between the proposed network and the deformable neural network, U-Net are provided in our study. Results show that more detailed vessels can be extracted by DUNet and it exhibits state-of-the-art performance for retinal vessel segmentation with a global accuracy of 0.9566/0.9641/0.9610/0.9651 and AUC of 0.9802/0.9832/0.9804/0.9831 on DRIVE, STARE, CHASE_DB1 and HRF respectively. Moreover, to show the generalization ability of the DUNet, we use another two retinal vessel data sets, i.e., WIDE and SYNTHE, to qualitatively and quantitatively analyze and compare with other methods. Extensive cross-training evaluations are used to further assess the extendibility of DUNet. The proposed method has the potential to be applied to the early diagnosis of diseases.

Place, publisher, year, edition, pages
Elsevier, 2019
Keywords
Retinal blood vessel, Segmentation, DUNet, U-Net, Deformable convolution
National Category
Medical Imaging
Identifiers
urn:nbn:se:liu:diva-157172 (URN)10.1016/j.knosys.2019.04.025 (DOI)000472687500013 ()2-s2.0-85065243868 (Scopus ID)
Note

Funding agencies: National Natural Science Foundation of China [61702361]; Science and Technology Program of Tianjin, China [16ZXHLGX00170]; National Key Technology R&D Program of China [2018YFB1701700]

Available from: 2019-06-01 Created: 2019-06-01 Last updated: 2025-02-09Bibliographically approved
Pham, T. (2019). Geostatistical Simulation of Medical Images for Data Augmentation in Deep Learning. IEEE Access, 7, 68752-68763
Open this publication in new window or tab >>Geostatistical Simulation of Medical Images for Data Augmentation in Deep Learning
2019 (English)In: IEEE Access, E-ISSN 2169-3536, Vol. 7, p. 68752-68763Article in journal (Refereed) Published
Abstract [en]

Data augmentation, which is the process for creating alternative copies of each sample in a small training data set, is important for extracting deep-learning features for medical image classification problems. However, data augmentation has not been well explored and existing methods are heuristic. In this paper, a geostatistical simulation of images is introduced as a data augmentation approach for extracting deep-learning features from medical images that are characterized with texture. The stochastic simulation procedure is to generate realizations of an image by modeling its spatial variability through the generation of multiple equiprobable stochastic realizations. The approach employed for the geostatistical simulation is based on the concepts of regionalized variables and kriging formulation to create multiple textural variations of medical images. Experimental results on classifying two medical-image data sets show that the use of geostatistical simulation for extracting deep-learning features with several popular pre-trained deep-learning models (AlexNet, ResNet-50, and GoogLeNet/Inception) provided better accuracy rates and more balanced results in terms of sensitivity and specificity than either with or without the implementation of conventional data augmentation. The proposed approach can be applied to other pre-trained networks as well as those without data pre-training for effective deep-feature extraction from medical images. The proposed application of the theory of geostatistics for medical image data augmentation in deep learning is original. The novelty of this approach can also be applied to many other types of data that are inherently textural. Geostatistical simulation opens a new door to the development of state-of-the-art artificial intelligence in feature extraction of medical images.

Place, publisher, year, edition, pages
Institute of Electrical and Electronics Engineers (IEEE), 2019
Keywords
Medical image features, sequential Gaussian simulation, geostatistics, deep learning, convolutional neural networks, classification.
National Category
Medical Imaging
Identifiers
urn:nbn:se:liu:diva-157171 (URN)10.1109/ACCESS.2019.2919678 (DOI)000471349000001 ()
Available from: 2019-06-01 Created: 2019-06-01 Last updated: 2025-02-09Bibliographically approved
Pham, T. (2019). Quantification of white matter lesions on brain MRI with 2D fuzzy weighted recurrence networks. In: : . Paper presented at 9th International IEEE/EMBS Conference on Neural Engineering (NER'19), San Francisco, CA, USA, 20-23 March 2019 (pp. 110-113). Institute of Electrical and Electronics Engineers (IEEE)
Open this publication in new window or tab >>Quantification of white matter lesions on brain MRI with 2D fuzzy weighted recurrence networks
2019 (English)Conference paper, Published paper (Refereed)
Abstract [en]

White matter lesions detected on magnetic resonance imaging scans of the brain have been hypothesized to have associations with the causes of several diseases. Accurate quantification of white matter lesions is important for the hypothesis validation. However, the clinical quantification is highly variable due to subjective opinions of different raters and is likely to compromise the reliability of the assessment. This paper introduces a new method of two-dimensional fuzzy weighted recurrence networks that can numerically express the quantity of white matter lesions. The results illustrate the promising application of the proposed method that offers as a useful computational tool in brain research.

Place, publisher, year, edition, pages
Institute of Electrical and Electronics Engineers (IEEE), 2019
National Category
Medical Imaging
Identifiers
urn:nbn:se:liu:diva-157173 (URN)10.1109/NER.2019.8717150 (DOI)000469933200028 ()9781538679210 (ISBN)9781538679227 (ISBN)
Conference
9th International IEEE/EMBS Conference on Neural Engineering (NER'19), San Francisco, CA, USA, 20-23 March 2019
Available from: 2019-06-01 Created: 2019-06-01 Last updated: 2025-02-09Bibliographically approved
Pham, T. (2019). Tensor decomposition of non-EEG physiological signals for visualization and recognition of human stress. In: ICBBT 2019: 2019 11TH INTERNATIONAL CONFERENCE ON BIOINFORMATICS AND BIOMEDICAL TECHNOLOGY: . Paper presented at 11th Int. Conf. Bioinformatics and Biomedical Technology, Stockholm, Sweden, May 29 - 31, 2019 (pp. 132-136). New York: ACM Publications
Open this publication in new window or tab >>Tensor decomposition of non-EEG physiological signals for visualization and recognition of human stress
2019 (English)In: ICBBT 2019: 2019 11TH INTERNATIONAL CONFERENCE ON BIOINFORMATICS AND BIOMEDICAL TECHNOLOGY, New York: ACM Publications, 2019, p. 132-136Conference paper, Published paper (Refereed)
Abstract [en]

Recognition of physical and mental responses to stress is important for the purpose of stress management to reduce its negative effects in health. Wearable technology, mainly using electroencephalogram (EEG), provides information such as tracking fitness activity, disease detection, and neurological states of individuals. However, the recording of EEG signals from a wearable device is inconvenient. This study introduces the application of tensor decomposition of non-EEG data for visualizing and tracking neurological status with implication to human stress recognition. Results obtained from testing the proposed method using a PhyioNet database show visualizations that can well separate four groups of neurological statuses obtained from twenty healthy subjects, and very high up to 100% classification of the neurological statuses. The investigation suggests the potential application of tensor decomposition for the analysis of physiological measurements collected from multiple sensors. The proposed study can significantly contribute to the advancement of wearable technology for human stress monitoring

Place, publisher, year, edition, pages
New York: ACM Publications, 2019
National Category
Other Medical Engineering Psychiatry
Identifiers
urn:nbn:se:liu:diva-159038 (URN)10.1145/3340074.3340096 (DOI)000519043000022 ()2-s2.0-85070558211 (Scopus ID)978-1-4503-6231-3 (ISBN)
Conference
11th Int. Conf. Bioinformatics and Biomedical Technology, Stockholm, Sweden, May 29 - 31, 2019
Available from: 2019-07-21 Created: 2019-07-21 Last updated: 2020-03-29Bibliographically approved
Cirillo, M. D., Mirdell, R., Sjöberg, F. & Pham, T. (2019). Time-Independent Prediction of Burn Depth using Deep Convolutional Neural Networks. Journal of Burn Care & Research, 40(6), 857-863
Open this publication in new window or tab >>Time-Independent Prediction of Burn Depth using Deep Convolutional Neural Networks
2019 (English)In: Journal of Burn Care & Research, ISSN 1559-047X, E-ISSN 1559-0488, Vol. 40, no 6, p. 857-863Article in journal (Refereed) Published
Abstract [en]

We present in this paper the application of deep convolutional neural networks, which are a state-of-the-art artificial intelligence (AI) approach in machine learning, for automated time-independent prediction of burn depth. Colour images of four types of burn depth injured in first few days, including normal skin and background, acquired by a TiVi camera were trained and tested with four pre-trained deep convolutional neural networks: VGG-16, GoogleNet, ResNet-50, and ResNet-101. In the end, the best 10-fold cross-validation results obtained from ResNet- 101 with an average, minimum, and maximum accuracy are 81.66%, 72.06% and 88.06%, respectively; and the average accuracy, sensitivity and specificity for the four different types of burn depth are 90.54%, 74.35% and 94.25%, respectively. The accuracy was compared to the clinical diagnosis obtained after the wound had healed. Hence, application of AI is very promising for prediction of burn depth and therefore can be a useful tool to help in guiding clinical decision and initial treatment of burn wounds.

Place, publisher, year, edition, pages
Oxford University Press, 2019
Keywords
Burn depth, time-independent prediction, deep convolutional neural network, artificial intelligence
National Category
Surgery Medical Imaging Other Clinical Medicine
Identifiers
urn:nbn:se:liu:diva-157386 (URN)10.1093/jbcr/irz103 (DOI)000495368300020 ()31187119 (PubMedID)
Note

Funding agencies: Analytic Imaging Diagnostic Arena (AIDA)

Available from: 2019-06-11 Created: 2019-06-11 Last updated: 2025-02-09Bibliographically approved
Liu, J., Pham, T., Yan, H. & Liang, Z. (2018). Fuzzy mixed-prototype clustering algorithm for microarray data analysis. Neurocomputing, 276, 42-54
Open this publication in new window or tab >>Fuzzy mixed-prototype clustering algorithm for microarray data analysis
2018 (English)In: Neurocomputing, ISSN 0925-2312, E-ISSN 1872-8286, Vol. 276, p. 42-54Article in journal (Refereed) Published
Abstract [en]

Being motivated by combining the advantages of hyperplane-based pattern analysis and fuzzy clustering techniques, we present in this paper a fuzzy mix-prototype (FMP) clustering for microarray data analysis. By integrating spherical and hyper-planar cluster prototypes, the FMP is capable of capturing latent data models with both spherical and non-spherical geometric structures. Our contributions of the paper can be summarized into three folds: first, the objective function of the FMP is formulated. Second, an iterative solution which minimizes the objective function under given constraints is derived. Third, the effectiveness of the proposed FMP is demonstrated through experiments on yeast and leukemia data sets.

Place, publisher, year, edition, pages
Elsevier, 2018
Keywords
FMP, Microarray data analysis, Fuzzy clustering
National Category
Computer Sciences
Identifiers
urn:nbn:se:liu:diva-141163 (URN)10.1016/j.neucom.2017.06.083 (DOI)000419222000005 ()
Available from: 2017-09-25 Created: 2017-09-25 Last updated: 2018-01-22Bibliographically approved
Pham, T. (2018). Pattern analysis and classification of blood oxygen saturation signals with nonlinear dynamics features. In: : . Paper presented at 2018 IEEE EMBS International Conference on Biomedical & Health Informatics (pp. 112-115). Institute of Electrical and Electronics Engineers (IEEE)
Open this publication in new window or tab >>Pattern analysis and classification of blood oxygen saturation signals with nonlinear dynamics features
2018 (English)Conference paper, Published paper (Refereed)
Abstract [en]

Pattern analysis of blood oxygen saturation is important for gaining insights into the cardiorespiratory control system, real-time monitoring during operations, identifying potential predictors for the diagnosis of disease severity, and improving the hospitalization of patients with critical chronic diseases. This paper investigates the use of nonlinear dynamics features for machine learning and classification of blood oxygen saturation signals in healthy young and healthy old subjects. The validation of the feature reliability for the signal variability analysis has a clinical implication for differentiating blood oxygen saturation in patients with respect to the particular influence of aging, when patient's data become available.

Place, publisher, year, edition, pages
Institute of Electrical and Electronics Engineers (IEEE), 2018
National Category
Other Medical Engineering
Identifiers
urn:nbn:se:liu:diva-147558 (URN)10.1109/BHI.2018.8333382 (DOI)978-1-5386-2405-0 (ISBN)978-1-5386-2406-7 (ISBN)
Conference
2018 IEEE EMBS International Conference on Biomedical & Health Informatics
Available from: 2018-04-26 Created: 2018-04-26 Last updated: 2018-05-18
Pham, T. (2018). Pattern analysis of computer keystroke time series in healthy control and early-stage Parkinson's disease subjects using fuzzy recurrence and scalable recurrence network features. Journal of Neuroscience Methods, 307, 128-130
Open this publication in new window or tab >>Pattern analysis of computer keystroke time series in healthy control and early-stage Parkinson's disease subjects using fuzzy recurrence and scalable recurrence network features
2018 (English)In: Journal of Neuroscience Methods, ISSN 0165-0270, E-ISSN 1872-678X, Vol. 307, p. 128-130Article in journal (Refereed) Published
Abstract [en]

A noncommutative algebra corresponding to the classical catenoid is introduced together with a differential calculus of derivations. We prove that there exists a unique metric and torsion-free connection that is compatible with the complex structure, and the curvature is explicitly calculated. A noncommutative analogue of the fact that the catenoid is a minimal surface is studied by constructing a Laplace operator from the connection and showing that the embedding coordinates are harmonic. Furthermore, an integral is defined and the total curvature is computed. Finally, classes of left and right modules are introduced together with constant curvature connections, and bimodule compatibility conditions are discussed in detail.

Place, publisher, year, edition, pages
Elsevier, 2018
Keywords
Noncommutative catenoid, Noncommutative Riemannian geometry, Noncommutative minimal surface, Noncommutative curvature
National Category
Other Medical Engineering
Identifiers
urn:nbn:se:liu:diva-148129 (URN)10.1016/j.jneumeth.2018.05.019 (DOI)000442055800020 ()29859213 (PubMedID)
Available from: 2018-05-30 Created: 2018-05-30 Last updated: 2018-10-04Bibliographically approved
Pham, T. (2017). Time-shift multiscale entropy analysis of physiological signals. Entropy, 19(6), Article ID 257.
Open this publication in new window or tab >>Time-shift multiscale entropy analysis of physiological signals
2017 (English)In: Entropy, E-ISSN 1099-4300, Vol. 19, no 6, article id 257Article in journal (Refereed) Published
Abstract [en]

Measures of predictability in physiological signals using entropy measures have been widely applied in many areas of research. Multiscale entropy expresses different levels of either approximate entropy or sample entropy by means of multiple factors for generating multiple time series, enabling the capture of more useful information than using a scalar value produced by the two entropy methods. This paper presents the use of different time shifts on various intervals of time series to discover different entropy patterns of the time series. Examples and experimental results using white noise, 1/ f noise, photoplethysmography, and electromyography signals suggest the validity and better performance of the proposed time-shift multiscale entropy analysis of physiological signals than the multiscale entropy.

Keywords
approximate entropy, sample entropy, multiscale entropy, higuchi’s fractal dimension, time shift, physiological signals
National Category
Other Medical Engineering
Identifiers
urn:nbn:se:liu:diva-138027 (URN)10.3390/e19060257 (DOI)000404454500020 ()
Note

Funding agencies: LiU Faculty of Science and Engineering

Available from: 2017-06-07 Created: 2017-06-07 Last updated: 2023-03-28Bibliographically approved
Organisations
Identifiers
ORCID iD: ORCID iD iconorcid.org/0000-0002-4255-5130

Search in DiVA

Show all publications

Profile pages

Personal homepage