1
|
Kent B, Rossa C. Electric impedance spectroscopy feature extraction for tissue classification with electrode embedded surgical needles through a modified forward stepwise method. Comput Biol Med 2021; 135:104522. [PMID: 34153792 DOI: 10.1016/j.compbiomed.2021.104522] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.7] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/07/2021] [Revised: 05/21/2021] [Accepted: 05/25/2021] [Indexed: 10/21/2022]
Abstract
There has been a growing interest in developing electric impedance sensing surgical tools for tissue identification during surgery. A key facet of this development is identifying distinct features that can be used to identify tissues from one another. This paper explores several feature extraction techniques and classification methods applied to electric impedance data. Furthermore, a modified forward stepwise method is proposed. The method introduces a scoring metric to help select features to add to the model, that is based off of the coefficient of variation and overlapping index from the feature's probability density functions for each of the classes. The proposed and existing methods were applied to spectral data measured at 23 frequencies, from 132 samples across 6 different tissues including ex-vivo bovine kidney, liver and muscle, poultry liver, as well as freshly excised canine testicle and ovary samples. These methods were able to successfully find impedance spectra features for the investigated biological tissues. The best predictive accuracy was with Boruta feature extraction and a Random Forest classifier but without significantly reducing the number of features in the classifier model. The proposed method was able to reduce the number of features in the model to an average of 5.8 features for all tested classifiers. These methods may have use in finding features to discriminate other tissue types, possibly to aid in targeting lesions in minimally invasive cancer treatment surgeries.
Collapse
Affiliation(s)
- B Kent
- Ontario Tech University, 2000 Simcoe Street North, Oshawa, ON L1G 0C5, Canada.
| | - C Rossa
- Ontario Tech University, 2000 Simcoe Street North, Oshawa, ON L1G 0C5, Canada.
| |
Collapse
|
2
|
Van Eeckhout A, Garcia-Caurel E, Ossikovski R, Lizana A, Rodríguez C, González-Arnay E, Campos J. Depolarization metric spaces for biological tissues classification. JOURNAL OF BIOPHOTONICS 2020; 13:e202000083. [PMID: 32406967 DOI: 10.1002/jbio.202000083] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Received: 03/09/2020] [Revised: 05/01/2020] [Accepted: 05/08/2020] [Indexed: 05/02/2023]
Abstract
Classification of tissues is an important problem in biomedicine. An efficient tissue classification protocol allows, for instance, the guided-recognition of structures through treated images or discriminating between healthy and unhealthy regions (e.g., early detection of cancer). In this framework, we study the potential of some polarimetric metrics, the so-called depolarization spaces, for the classification of biological tissues. The analysis is performed using 120 biological ex vivo samples of three different tissues types. Based on these data collection, we provide for the first time a comparison between these depolarization spaces, as well as with most commonly used depolarization metrics, in terms of biological samples discrimination. The results illustrate the way to determine the set of depolarization metrics which optimizes tissue classification efficiencies. In that sense, the results show the interest of the method which is general, and which can be applied to study multiple types of biological samples, including of course human tissues. The latter can be useful for instance, to improve and to boost applications related to optical biopsy.
Collapse
Affiliation(s)
- Albert Van Eeckhout
- Grup d'Òptica, Physics Department, Universitat Autònoma de Barcelona, Bellaterra, Spain
| | - Enric Garcia-Caurel
- LPICM, CNRS, École Polytechnique, Université Paris-Saclay, Palaiseau, France
| | - Razvigor Ossikovski
- LPICM, CNRS, École Polytechnique, Université Paris-Saclay, Palaiseau, France
| | - Angel Lizana
- Grup d'Òptica, Physics Department, Universitat Autònoma de Barcelona, Bellaterra, Spain
| | - Carla Rodríguez
- Grup d'Òptica, Physics Department, Universitat Autònoma de Barcelona, Bellaterra, Spain
| | - Emilio González-Arnay
- Departamento de Anatomía, Histología y Neurociencia, Universidad Autónoma de Madrid, Madrid, Spain
- Servicio de Anatomía Patológica, Hospital Universitario de Canarias, Santa Cruz de Tenerife, Spain
| | - Juan Campos
- Grup d'Òptica, Physics Department, Universitat Autònoma de Barcelona, Bellaterra, Spain
| |
Collapse
|
3
|
Kim TH, Choi A, Heo HM, Kim K, Lee K, Mun JH. Machine learning-based pre-impact fall detection model to discriminate various types of fall. J Biomech Eng 2019; 141:2730876. [PMID: 30968932 DOI: 10.1115/1.4043449] [Citation(s) in RCA: 16] [Impact Index Per Article: 3.2] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/06/2018] [Indexed: 11/08/2022]
Abstract
Preimpact fall detection can send alarm service faster to reduce long-lie conditions and decrease the risk of hospitalization. Detecting various types of fall to determine the impact site or direction prior to impact is important because it increases the chance of decreasing the incidence or severity of fall-related injuries. In this study, a robust preimpact fall detection model was developed to classify various activities and falls as multi-class and its performance was compared with the performance of previous developed models. Twelve healthy subjects participated in this study. All subjects were asked to place an inertial measuring unit module by fixing on a belt near the left iliac crest to collect accelerometer data for each activity. Our novel proposed model consists of feature calculation and infinite latent feature selection algorithm, auto labeling of activities, application of machine learning classifiers for discrete and continuous time series data. Nine machine-learning classifiers were applied to detect falls prior to impact and derive final detection results by sorting the classifier. Our model showed the highest classification accuracy. Results for the proposed model that could classify as multi-class showed significantly higher average classification accuracy of 99.57 ± 0.01% for discrete data-based classifiers and 99.84 ± 0.02% for continuous time series-based classifiers than previous models (p < 0.01). In the future, multi-class preimpact fall detection models can be applied to fall protector devices by detecting various activities for sending alerts or immediate feedback reactions to prevent falls.
Collapse
Affiliation(s)
- Tae Hyong Kim
- Department of Biomechatronic Engineering, College of Biotechnology and Bioengineering, Sungkyunkwan University, Republic of Korea, 2066 Seoburo, Jangangu, Suwon, Gyeonggi, 16419, Republic of Korea
| | - Ahnryul Choi
- Department of Biomechatronic Engineering, College of Biotechnology and Bioengineering, Sungkyunkwan University, Republic of Korea, 2066 Seoburo, Jangangu, Suwon, Gyeonggi, 16419, Republic of Korea; Department of Biomedical Engineering, College of Medical Convergence, Catholic Kwandong University, Republic of Korea, 24, Beomil-ro 579 beon-gill, Gangneung, Gangwon, Republic of Korea
| | - Hyun Mu Heo
- Department of Biomechatronic Engineering, College of Biotechnology and Bioengineering, Sungkyunkwan University, Republic of Korea, 2066 Seoburo, Jangangu, Suwon, Gyeonggi, 16419, Republic of Korea
| | - Kyungran Kim
- Agricultural Health and Safety Division, Rural Development Administration, Republic of Korea, 300 Nongsaengmyeong-ro, Wansan-gu, Jeonju, Jeollabuk 54875, Republic of Korea
| | - Kyungsuk Lee
- Agricultural Health and Safety Division, Rural Development Administration, Republic of Korea, 300 Nongsaengmyeong-ro, Wansan-gu, Jeonju, Jeollabuk 54875, Republic of Korea
| | - Joung Hwan Mun
- Department of Biomechatronic Engineering, College of Biotechnology and Bioengineering, Sungkyunkwan University, Republic of Korea, 2066 Seoburo, Jangangu, Suwon, Gyeonggi, 16419, Republic of Korea, Tel: +82-31-290-7827, Fax: +82-31-290-7830
| |
Collapse
|
4
|
Joo SB, Oh SE, Mun JH. Improving the ground reaction force prediction accuracy using one-axis plantar pressure: Expansion of input variable for neural network. J Biomech 2016; 49:3153-3161. [PMID: 27515436 DOI: 10.1016/j.jbiomech.2016.07.029] [Citation(s) in RCA: 5] [Impact Index Per Article: 0.6] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/05/2015] [Revised: 07/25/2016] [Accepted: 07/26/2016] [Indexed: 10/21/2022]
Abstract
In this study, we describe a method to predict 6-axis ground reaction forces based solely on plantar pressure (PP) data obtained from insole type measurement devices free of space limitations. Because only vertical force is calculable from PP data, a wavelet neural network derived from a non-linear mapping function was used to obtain 3-axis ground reaction force in medial-lateral (GRFML), anterior-posterior (GRFAP) and vertical (GRFV) and 3-axis ground reaction moment in sagittal (GRFS), frontal (GRFF) and transverse (GRFT) data for the remaining axes and planes. As the prediction performance of nonlinear models depends strongly on input variables, in this study, three input variables - accumulated PP with respect to time, center of pressure (COP) pattern, and measurements of the opposite foot, which are calculable only with a PP device - were considered in order to improve prediction performance. To conduct this study, the golf swing motions of 80 subjects were characterized as unilateral movement and GRF patterns as functions of individual characteristics. The prediction model was verified with 5-fold cross-validation utilizing the measured values of two force plates. As a result, prediction model (correlation coefficient, r=0.73-0.97) utilized accumulated PP and PP data of the opposite foot and showed the highest prediction accuracy in left-foot GRFV, GRMF, GRMT and right-foot GRFAP, GRFML, GRMF, GRMT. Likewise, another prediction model (r=0.83-0.98) utilized accumulated PP and COP patterns as input and showed the best accuracy in left-foot GRFAP, GRFML, GRMS and right-foot GRFV, GRMS. New methods based on the findings of the present study are expected to help resolve problems such as spatial limitation and limited analyzable motions in existing GRF measurement processes.
Collapse
Affiliation(s)
- Su-Bin Joo
- Department of Bio-Mechatronic Engineering, College of Biotechnology and Bioengineering, Sungkyunkwan University, 300 Chunchun, Jangan, Suwon, Gyeonggi 440-746, Republic of Korea
| | - Seung Eel Oh
- Department of Bio-Mechatronic Engineering, College of Biotechnology and Bioengineering, Sungkyunkwan University, 300 Chunchun, Jangan, Suwon, Gyeonggi 440-746, Republic of Korea; Research Group of Smart Food Distribution System, Korea Food Research Institute, Seongnam, Gyeonggi 463-746, Republic of Korea
| | - Joung Hwan Mun
- Department of Bio-Mechatronic Engineering, College of Biotechnology and Bioengineering, Sungkyunkwan University, 300 Chunchun, Jangan, Suwon, Gyeonggi 440-746, Republic of Korea.
| |
Collapse
|