Predicting the Healing of Lower Extremity Fractures Using Wearable Ground Reaction Force Sensors and Machine Learning
Abstract
:1. Introduction
2. Materials and Methods
2.1. Study Design
2.2. Data Reduction
2.2.1. Step Extraction and Discretization
2.2.2. Random Sampling and Feature Selection
2.3. Logistic Regression Classifier
2.4. Examining Trends in Selected Features
3. Results
3.1. Study Design
3.2. Data Reduction—Step Extraction and Discretization
3.3. Data Reduction—Random Sampling and Feature Selection
3.4. Logistic Regression Classifier
3.5. Examining Trends in Selected Features
4. Discussion
5. Conclusions
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Conflicts of Interest
References
- Mock, C.; Cherian, M.N. The global burden of musculoskeletal injuries: Challenges and solutions. Clin. Orthop. Relat. Res. 2008, 466, 2306–2316. [Google Scholar] [CrossRef] [PubMed]
- Court-Brown, C.M.; Caesar, B. Epidemiology of adult fractures: A review. Injury 2006, 37, 691–697. [Google Scholar] [CrossRef] [PubMed]
- Larsen, P.; Elsoe, R.; Hansen, S.H.; Graven-Nielsen, T.; Laessoe, U.; Rasmussen, S. Incidence and epidemiology of tibial shaft fractures. Injury 2015, 46, 746–750. [Google Scholar] [CrossRef] [PubMed]
- Hoyt, B.W.; Pavey, G.J.; Pasquina, P.F.; Potter, B.K. Rehabilitation of Lower Extremity Trauma: A Review of Principles and Military Perspective on Future Directions. Curr. Trauma Rep. 2015, 1, 50–60. [Google Scholar] [CrossRef]
- Lin, C.W.C.; Donkers, N.A.; Refshauge, K.M.; Beckenkamp, P.R.; Khera, K.; Moseley, A.M. Rehabilitation for ankle fractures in adults. Cochrane Database Syst. Rev. 2012, 11, CD005595. [Google Scholar] [PubMed]
- Martin, R.L.; McPoil, T.G. Reliability of ankle goniometric measurements: A literature review. J. Am. Podiatr. Med. Assoc. 2005, 95, 564–572. [Google Scholar] [CrossRef]
- Blokhuis, T.; De Bruine, J.; Bramer, J.; Den Boer, F.; Bakker, F.; Patka, P.; Haarman, H.T.M.; Manoliu, R. The reliability of plain radiography in experimental fracture healing. Skelet. Radiol. 2001, 30, 151–156. [Google Scholar] [CrossRef]
- Chang, E.M.; Gillespie, E.F.; Shaverdian, N. Truthfulness in patient-reported outcomes: Factors affecting patients’ responses and impact on data quality. Patient Relat. Outcome Meas. 2019, 10, 171–186. [Google Scholar] [CrossRef]
- MacKenzie, E.J.; Bosse, M.J. Factors influencing outcome following limb-threatening lower limb trauma: Lessons learned from the Lower Extremity Assessment Project (LEAP). J. Am. Acad. Orthop. Surg. 2006, 14, S205–S210. [Google Scholar] [CrossRef]
- McCarthy, M.L.; MacKenzie, E.J.; Edwin, D.; Bosse, M.J.; Castillo, R.C.; Starr, A.; Kellam, J.F.; Burgess, A.R.; Webb, L.X.; Swiontkowski, M.F. Psychological distress associated with severe lower-limb injury. JBJS 2003, 85, 1689–1697. [Google Scholar] [CrossRef] [PubMed]
- Sliepen, M.; Lipperts, M.; Tjur, M.; Mechlenburg, I. Use of accelerometer-based activity monitoring in orthopaedics: Benefits, impact and practical considerations. EFORT Open Rev. 2019, 4, 678. [Google Scholar] [CrossRef] [PubMed]
- Swart, E.; Peindl, R.; Zheng, N.; Habet, N.; Churchill, C.; Ruder, J.A.; Seymour, R.; Karunakar, M.; Kellam, J.; Sims, S. Electronically augmented gait abnormality assessment following lower extremity trauma. OTA Int. 2019, 2, e032. [Google Scholar] [CrossRef] [PubMed]
- Majumder, S.; Mondal, T.; Deen, M.J. Wearable Sensors for Remote Health Monitoring. Sensors 2017, 17, 130. [Google Scholar] [CrossRef] [PubMed]
- Redmond, S.J.; Lovell, N.H.; Yang, G.Z.; Horsch, A.; Lukowicz, P.; Murrugarra, L.; Marschollek, M. What Does Big Data Mean for Wearable Sensor Systems? Contribution of the IMIA Wearable Sensors in Healthcare WG. Yearb. Med. Inform. 2014, 9, 135–142. [Google Scholar] [CrossRef] [PubMed] [PubMed Central]
- Vijayan, V.; Connolly, J.P.; Condell, J.; McKelvey, N.; Gardiner, P. Review of Wearable Devices and Data Collection Considerations for Connected Health. Sensors 2021, 21, 5589. [Google Scholar] [CrossRef]
- Yanardag, M.; Şimşek, T.T.; Yanardag, F. Exploring the Relationship of Pain, Balance, Gait Function, and Quality of Life in Older Adults with Hip and Knee Pain. Pain Manag. Nurs. 2021, 22, 503–508. [Google Scholar] [CrossRef] [PubMed]
- Van Hoeve, S.; Houben, M.; Verbruggen, J.P.A.M.; Willems, P.; Meijer, K.; Poeze, M. Gait analysis related to functional outcome in patients operated for ankle fractures. J. Orthop. Res. 2019, 37, 1658–1666. [Google Scholar] [CrossRef]
- Hamill, J.; Knutzen, K.M. Biomechanical Basis of Human Movement; Lippincott Williams & Wilkins: Philadelphia, PA, USA, 2006. [Google Scholar]
- Marasović, T.; Cecić, M.; Zanchi, V. Analysis and interpretation of ground reaction forces in normal gait. WSEAS Trans. Syst. 2009, 8, 1105–1114. [Google Scholar]
- Zadpoor, A.A.; Nikooyan, A.A. The relationship between lower-extremity stress fractures and the ground reaction force: A systematic review. Clin. Biomech. 2011, 26, 23–28. [Google Scholar] [CrossRef]
- McCrory, J.L.; White, S.C.; Lifeso, R.M. Vertical ground reaction forces: Objective measures of gait following hip arthroplasty. Gait Posture 2001, 14, 104–109. [Google Scholar] [CrossRef]
- Lee, C.E.; Simmonds, M.J.; Etnyre, B.R.; Morris, G.S. Influence of pain distribution on gait characteristics in patients with low back pain: Part 1: Vertical ground reaction force. Spine 2007, 32, 1329–1336. [Google Scholar] [CrossRef] [PubMed]
- Madadi-Shad, M.; Jafarnezhadgero, A.; Zago, M.; Granacher, U. Effects of varus knee alignment on gait biomechanics and lower limb muscle activity in boys: A cross sectional study. Gait Posture 2019, 72, 69–75. [Google Scholar] [CrossRef] [PubMed]
- Obermeyer, Z.; Emanuel, E.J. Predicting the Future—Big Data, Machine Learning, and Clinical Medicine. N. Engl. J. Med. 2016, 375, 1216–1219. [Google Scholar] [CrossRef] [PubMed]
- Intelligent health data analytics: A convergence of artificial intelligence and big data. In Healthcare Management Forum; Abidi, S.S.R.; Abidi, S.R. (Eds.) SAGE Publications: Los Angeles, CA, USA, 2019. [Google Scholar]
- Loyola-Gonzalez, O. Black-box vs. white-box: Understanding their advantages and weaknesses from a practical point of view. IEEE Access 2019, 7, 154096–154113. [Google Scholar] [CrossRef]
- North, K.; Kubiak, E.N.; Rothberg, D.L.; Lajevardi-Khosh, A.; Petelenz, T.J.; Hitchcock, R.W.; Stuart, A.R. Longitudinal monitoring of patient limb loading throughout ankle fracture rehabilitation using an insole load monitoring system: A case series. Curr. Orthop. Pract. 2017, 28, 223–230. [Google Scholar] [CrossRef]
- Lajevardi-Khosh, A.; Stuart, A.; Ackerman, M.; Rothberg, D.; Kubiak, E.; Petelenz, T.; Hitchcock, R. Characterization of compliance to weight-bearing protocols and patient weight-bearing behavior during the recovery period in lower extremity fractures: A pilot study. Curr. Orthop. Pract. 2019, 30, 395–402. [Google Scholar] [CrossRef]
- North, K.; Simpson, G.M.; Stuart, A.R.; Kubiak, E.N.; Petelenz, T.J.; Hitchcock, R.W.; Rothberg, D.L.; Cizik, A.M. Early postoperative step count and walking time have greater impact on lower limb fracture outcomes than load-bearing metrics. Injury 2023, 54, 110756. [Google Scholar] [CrossRef] [PubMed]
- North, K.; Kubiak, E.N.; Hitchcock, R.W.; Petelenz, T.J. Load Monitoring System for Partial Weight Bearing Therapy for rehabilitation of lower extremity fractures. In Proceedings of the 2013 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Osaka, Japan, 3–7 July 2013. [Google Scholar]
- Tudor-Locke, C.; Han, H.; Aguiar, E.J.; Barreira, T.V.; Schuna, J.M., Jr.; Kang, M.; Rowe, D.A. How fast is fast enough? Walking cadence (steps/min) as a practical estimate of intensity in adults: A narrative review. Br. J. Sports Med. 2018, 52, 776–788. [Google Scholar] [CrossRef]
- Tudor-Locke, C.; Rowe, D.A. Using cadence to study free-living ambulatory behaviour. Sports Med. 2012, 42, 381–398. [Google Scholar] [CrossRef]
- Midi, H.; Bagheri, A. Robust multicollinearity diagnostic measure in collinear data set. In Proceedings of the 4th International Conference on Applied Mathematics, Simulation, Modeling, Corfu Island, Greece, 22–25 July 2010. [Google Scholar]
- Wong, T.-T. Performance evaluation of classification algorithms by k-fold and leave-one-out cross validation. Pattern Recognit. 2015, 48, 2839–2846. [Google Scholar] [CrossRef]
- Dreiseitl, S.; Ohno-Machado, L. Logistic regression and artificial neural network classification models: A methodology review. J. Biomed. Inform. 2002, 35, 352–359. [Google Scholar] [CrossRef]
- Muniz, A.; Liu, H.; Lyons, K.; Pahwa, R.; Liu, W.; Nobre, F.; Nadal, J. Comparison among probabilistic neural network, support vector machine and logistic regression for evaluating the effect of subthalamic stimulation in Parkinson disease on ground reaction force during gait. J. Biomech. 2010, 43, 720–726. [Google Scholar] [CrossRef] [PubMed]
- Park, S.J.; Hussain, I.; Hong, S.; Kim, D.; Park, H.; Benjamin, H.C.M. Real-time gait monitoring system for consumer stroke prediction service. In Proceedings of the 2020 IEEE International Conference on Consumer Electronics (ICCE), Las Vegas, NV, USA, 4–6 January 2020. [Google Scholar]
- Fonti, V.; Belitser, E. Feature selection using lasso. VU Amst. Res. Pap. Bus. Anal. 2017, 30, 1–25. [Google Scholar]
- Camargo, J.; Young, A. Feature Selection and Non-Linear Classifiers: Effects on Simultaneous Motion Recognition in Upper Limb. IEEE Trans. Neural Syst. Rehabil. Eng. 2019, 27, 743–750. [Google Scholar] [CrossRef] [PubMed]
- Braun, B.J.; Veith, N.T.; Rollmann, M.; Orth, M.; Fritz, T.; Herath, S.C.; Holstein, J.H.; Pohlemann, T. Weight-bearing recommendations after operative fracture treatment—Fact or fiction? Gait results with and feasibility of a dynamic, continuous pedobarography insole. Int. Orthop. 2017, 41, 1507–1512. [Google Scholar] [CrossRef]
- Eickhoff, A.M.; Cintean, R.; Fiedler, C.; Gebhard, F.; Schütze, K.; Richter, P.H. Analysis of partial weight bearing after surgical treatment in patients with injuries of the lower extremity. Orthop. Surg. 2022, 142, 77–81. [Google Scholar] [CrossRef]
- Warmerdam, E.; Orth, M.; Pohlemann, T.; Ganse, B. Gait Analysis to Monitor Fracture Healing of the Lower Leg. Bioengineering 2023, 10, 255. [Google Scholar] [CrossRef] [PubMed]
- Antipova, D.; Eadie, L.; Macaden, A.; Wilson, P. Diagnostic accuracy of clinical tools for assessment of acute stroke: A systematic review. BMC Emerg. Med. 2019, 19, 49. [Google Scholar] [CrossRef]
- Moenen, F.C.J.I.; Nelemans, P.J.; Schols, S.E.M.; Schouten, H.C.; Henskens, Y.M.C.; Beckers, E.A.M. The diagnostic accuracy of bleeding assessment tools for the identification of patients with mild bleeding disorders: A systematic review. Haemophilia 2018, 24, 525–535. [Google Scholar] [CrossRef]
- Haroon, S.; Jordan, R.; Takwoingi, Y.; Adab, P. Diagnostic accuracy of screening tests for COPD: A systematic review and meta-analysis. BMJ Open 2015, 5, e008133. [Google Scholar] [CrossRef]
- Berg, W.A.; Gutierrez, L.; NessAiver, M.S.; Carter, W.B.; Bhargavan, M.; Lewis, R.S.; Ioffe, O.B. Diagnostic accuracy of mammography, clinical examination, US, and MR imaging in preoperative assessment of breast cancer. Radiology 2004, 233, 830–849. [Google Scholar] [CrossRef] [PubMed]
- Lavallee, D.C.; Lee, J.R.; Austin, E.; Bloch, R.; Lawrence, S.O.; McCall, D.; Munson, S.A.; Nery-Hurwit, M.B.; Amtmann, D. mHealth and patient generated health data: Stakeholder perspectives on opportunities and barriers for transforming healthcare. Mhealth 2020, 6. [Google Scholar] [CrossRef] [PubMed]
Feature | FS Model | VIF | Importance Score | |||||
---|---|---|---|---|---|---|---|---|
DHA | DH4 | DH3 | DH2 | DH1 | DH0 | |||
OTA complication | Both | 1.67 | −3.20 | 4.59 | −2.78 | −1.76 | −1.65 | 1.80 |
OTA location | Both | 1.44 | 2.38 | −5.89 | 2.79 | 1.48 | 0.95 | −0.78 |
Mean of the medial sensor | DT | 1.60 | −1.25 | 3.74 | −1.30 | −1.23 | −0.67 | 0.49 |
Day | Both | 1.23 | 2.99 | −0.37 | 0.25 | 1.24 | 2.32 | −2.99 |
Variance of medial sensor | DT | 2.44 | 1.60 | −1.68 | 0.86 | 1.46 | 1.04 | −0.96 |
Heel to medial ratio | Lasso | 5.19 | −1.84 | 0.46 | −0.32 | 0.03 | −1.66 | 2.28 |
Mean of Diff_HF | Lasso | 1.13 | 1.33 | 0.88 | 0.28 | 0.62 | 1.22 | −1.46 |
Gender | Both | 2.70 | −0.79 | 2.41 | −0.63 | 0.00 | −0.48 | 0.83 |
Variance of the total waveform | Lasso | 2.86 | 0.20 | −3.02 | 0.55 | −0.37 | −0.08 | −0.13 |
Lateral to the medial ratio | DT | 4.81 | −1.00 | 0.14 | −1.11 | −2.43 | 0.02 | −0.04 |
Location of the maximum on Diff_HF | Lasso | 3.55 | −0.29 | 0.01 | 0.04 | −0.12 | −0.29 | 0.41 |
Feature | FS Model | VIF | Importance Score | |||||
---|---|---|---|---|---|---|---|---|
DHA | DH4 | DH3 | DH2 | DH1 | DH0 | |||
OTA complication | Both | 1.67 | −3.20 | 4.59 | −2.78 | −1.76 | −1.65 | 1.80 |
OTA location | Both | 1.44 | 2.38 | −5.89 | 2.79 | 1.48 | 0.95 | −0.78 |
Mean of the medial sensor | DT | 1.60 | −1.25 | 3.74 | −1.30 | −1.23 | −0.67 | 0.49 |
Day | Both | 1.23 | 2.99 | −0.37 | 0.25 | 1.24 | 2.32 | −2.99 |
Variance of the medial sensor | DT | 2.44 | 1.60 | −1.68 | 0.86 | 1.46 | 1.04 | −0.96 |
Heel to medial ratio | Lasso | 5.19 | −1.84 | 0.46 | −0.32 | 0.03 | −1.66 | 2.28 |
Mean of Diff_HF | Lasso | 1.13 | 1.33 | 0.88 | 0.28 | 0.62 | 1.22 | −1.46 |
Gender | Both | 2.70 | −0.79 | 2.41 | −0.63 | 0.00 | −0.48 | 0.83 |
Variance of the total waveform | Lasso | 2.86 | 0.20 | −3.02 | 0.55 | −0.37 | −0.08 | −0.13 |
Lateral to the medial ratio | DT | 4.81 | −1.00 | 0.14 | −1.11 | −2.43 | 0.02 | −0.04 |
Location of the maximum on Diff_HF | Lasso | 3.55 | −0.29 | 0.01 | 0.04 | −0.12 | −0.29 | 0.41 |
Leave one patient out | Precision | Recall | F1-score | Accuracy | |
Minimum | 75.2% | 74.4% | 74.2% | 74.4% | |
Mean | 76.7% | 76.0% | 75.9% | 76.0% | |
Maximum | 78.9% | 78.5% | 78.1% | 78.5% | |
10 fold cross-validation | Precision | Recall | F1-score | Accuracy | |
Minimum | 75.0% | 74.7% | 74.4% | 74.7% | |
Mean | 76.1% | 75.6% | 75.4% | 75.6% | |
Maximum | 77.5% | 76.9% | 76.7% | 76.9% |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
North, K.; Simpson, G.; Geiger, W.; Cizik, A.; Rothberg, D.; Hitchcock, R. Predicting the Healing of Lower Extremity Fractures Using Wearable Ground Reaction Force Sensors and Machine Learning. Sensors 2024, 24, 5321. https://doi.org/10.3390/s24165321
North K, Simpson G, Geiger W, Cizik A, Rothberg D, Hitchcock R. Predicting the Healing of Lower Extremity Fractures Using Wearable Ground Reaction Force Sensors and Machine Learning. Sensors. 2024; 24(16):5321. https://doi.org/10.3390/s24165321
Chicago/Turabian StyleNorth, Kylee, Grange Simpson, Walt Geiger, Amy Cizik, David Rothberg, and Robert Hitchcock. 2024. "Predicting the Healing of Lower Extremity Fractures Using Wearable Ground Reaction Force Sensors and Machine Learning" Sensors 24, no. 16: 5321. https://doi.org/10.3390/s24165321