Michelle P. Kwon
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Michelle P. Kwon has a total of 13 co-authors. Here are the top co-authors by number of shared publications:
- Josh R. Baxter
University of Pennsylvania
3 shared publications - Andy K. Smith
University of Delaware
1 shared publication Casey J. Humbyrd
University of Pennsylvania
1 shared publication- Casey Jo Humbyrd
Department of Orthopaedic Surgery, University of Pennsylvania, Philadelphia, PA, USA
1 shared publication - Celeste Durnwald
Department of Obstetrics and Gynecology, University of Pennsylvania, 1 West Gates, 3400 Spruce Street, Philadelphia, PA 19104, USA
1 shared publication - Devyn Russo
Department of Orthopaedic Surgery, University of Pennsylvania, 372A Stemmler Hall, Hamilton Walk, Philadelphia, PA 19104, USA
1 shared publication
Latest Publications
- 10 July 2026
- Published by Wolters Kluwer Health in Clinical Orthopaedics and Related Research
Foot Shape Changes During Pregnancy and Postpartum Are Associated With Patient Pain and Function
Background: Foot pain and symptoms affect 42% of pregnant women and extend beyond gestation, but the associations between foot morphology and these symptoms are not well understood. We sought to identify changes in foot shape throughout pregnancy and postpartum, using a novel multiplane digital camera–based system and to determine whether these alterations are associated with clinically important changes in the Patient-Reported Outcomes Measurement Information System (PROMIS) Pain Interference and Physical Function scores. Questions/purposes: (1) How does foot width, length, dorsum height, Arch Height Index, and medial foot area change throughout pregnancy and postpartum? (2) Are there relationships between foot shape and changes in foot shape over time with clinically important changes in PROMIS Pain Interference and Physical Function scores during pregnancy and postpartum? Methods: Between December 2023 and November 2024, we recruited patients scheduled for an obstetrics visit at Penn Medicine clinics in the Philadelphia, PA, USA, area who were between the gestational ages of 6 to 15 weeks, older than 18 years of age, had a BMI less than 45, had no history of connective tissue disorders or trauma to the lower extremities, and could understand and provide written consent. Of 2067 contacted patients, 40 consented and scheduled a laboratory visit; 78% (31 of 40) had complete datasets and were available for analysis. Pregnant participants (mean ± SD age of 35 ± 4 years, mean ± SD BMI of 68.8 ± 11.8 kg) visited the laboratory between 9 and 15 weeks (early pregnancy), 23 and 27 weeks (mid-pregnancy), and 34 and 39 weeks (late pregnancy) of gestation as well as between 9 and 15 weeks postpartum for foot shape assessment and completion of PROMIS scores. We developed a multiplane digital imaging approach designed to capture the top and medial views of the foot simultaneously to allow rapid quantification of foot shape while eliminating exposure to ionizing radiation. We validated this system by longitudinally monitoring foot shape in a control group of 18 nonpregnant participants (mean ± SD age of 28 ± 4 years, mean ± SD BMI of 62.7 ± 6.4 kg) assessed 12 weeks apart. This control group included female patients who had no previous pregnancies, were older than 18 years of age, had a BMI less than 45, had no history of connective tissue disorders or trauma to the lower extremities that resulted in medical intervention, and could provide written consent. We collected PROMIS Pain Interference and Physical Function score assessments during pregnancy and postpartum. Both scores are standardized T-scores and range from 20 to 80, with higher Pain Interference and lower Physical Function scores indicating worse outcomes. Minimum clinically important differences for pain interference and physical function in foot and ankle orthopaedics range between 5.5 and 9.8 T-score points; therefore, we used a conservative threshold of > 10, which represents 1 SD, to identify clinically important changes in pain interference and physical function during pregnancy. Results: We found that foot shape changed from early to late pregnancy, with median (range) increases in left foot width (9.14 cm [8.42 to 11.50] versus 9.30 cm [8.55 to 10.57], median difference +0.21 cm [95% CI 0.06 to 0.25]; p = 0.02), left dorsum height (6.22 cm [5.52 to 7.12] versus 6.33 cm [5.67 to 7.78], median difference +0.17 cm [95% CI 0.01 to 0.46]; p = 0.03), and right dorsum height (6.30 cm [5.51 to 7.58] versus 6.50 cm [5.52 to 7.64], median difference +0.23 cm [95% CI 0.13 to 0.38]; p = 0.008). There were mean ± SD increases in left foot length (24.37 ± 1.24 cm versus 24.66 ± 1.12 cm, mean difference +0.29 cm [95% CI 0.09 to 0.49]; p = 0.03), left medial foot area (83.08 ± 7.12 cm 2 versus 87.30 ± 7.15 cm 2 , mean difference +4.22 cm 2 [95% CI 2.19 to 6.25]; p = 0.001), and right medial foot area (84.02 ± 7.71 cm 2 versus 88.10 ± 8.53 cm 2 , mean difference +4.08 cm 2 [95% CI 2.21 to 5.94]; p < 0.001). In addition, median (range) left foot width increased from early pregnancy to the postpartum period (9.14 cm [8.42 to 11.50] versus 9.42 cm [8.29 to 10.77], median difference +0.10 cm [95% CI 0.09 to 0.37]; p = 0.01). We found associations between foot shape and patient-reported pain during pregnancy. Pregnant participants with clinically important increases in pain interference from early to late pregnancy had increased right medial foot area during late pregnancy (median [range] 11.37 cm 2 /shoe size [9.72 to 15.97], median difference +1.02 cm 2 /shoe size [95% CI 0.42 to 1.73]; p = 0.005) and postpartum (11.26 cm 2 /shoe size [9.88 to 15.47], median difference +0.91 cm 2 /shoe size [95% CI 0.04 to 2.57]; p = 0.01) compared to the control group of nonpregnant participants (10.35 cm 2 /shoe size [8.76 to 11.54]). Pregnant participants with clinically important increases in pain interference from early to late pregnancy also demonstrated an increase in right foot length from early to late pregnancy when compared with nonpregnant participants who had no changes in pain interference (0.49 ± 0.59 cm versus -0.14 ± 0.49 cm, mean difference 0.63 cm [95% CI 0.21 to 1.06]; p = 0.005). Pregnant participants with clinically important declines in physical function had a greater increase in right medial foot area from early to late pregnancy than pregnant participants without a physical function decline (6.62 ± 4.94 cm 2 versus 1.98 ± 4.00 cm 2 , mean difference 4.64 cm 2 [95% CI 1.14 to 4.64]; p = 0.01). Conclusion: This study identified potential morphological risk factors for musculoskeletal pain and reduced function in pregnant patients. Patient-reported pain and function were predominantly associated with pregnancy-induced lower extremity edema. These findings have potential clinical impact to guide interventions (such as compression socks, massage, or foot elevation) aimed at managing foot changes during pregnancy to reduce musculoskeletal pain and disability. Level of Evidence Level III, prognostic study.
- 10 July 2026
- Published by Wolters Kluwer Health in Clinical Orthopaedics and Related Research
Foot Shape Changes During Pregnancy and Postpartum Are Associated With Patient Pain and Function
Background: Foot pain and symptoms affect 42% of pregnant women and extend beyond gestation, but the associations between foot morphology and these symptoms are not well understood. We sought to identify changes in foot shape throughout pregnancy and postpartum, using a novel multiplane digital camera–based system and to determine whether these alterations are associated with clinically important changes in the Patient-Reported Outcomes Measurement Information System (PROMIS) Pain Interference and Physical Function scores. Questions/purposes: (1) How does foot width, length, dorsum height, Arch Height Index, and medial foot area change throughout pregnancy and postpartum? (2) Are there relationships between foot shape and changes in foot shape over time with clinically important changes in PROMIS Pain Interference and Physical Function scores during pregnancy and postpartum? Methods: Between December 2023 and November 2024, we recruited patients scheduled for an obstetrics visit at Penn Medicine clinics in the Philadelphia, PA, USA, area who were between the gestational ages of 6 to 15 weeks, older than 18 years of age, had a BMI less than 45, had no history of connective tissue disorders or trauma to the lower extremities, and could understand and provide written consent. Of 2067 contacted patients, 40 consented and scheduled a laboratory visit; 78% (31 of 40) had complete datasets and were available for analysis. Pregnant participants (mean ± SD age of 35 ± 4 years, mean ± SD BMI of 68.8 ± 11.8 kg) visited the laboratory between 9 and 15 weeks (early pregnancy), 23 and 27 weeks (mid-pregnancy), and 34 and 39 weeks (late pregnancy) of gestation as well as between 9 and 15 weeks postpartum for foot shape assessment and completion of PROMIS scores. We developed a multiplane digital imaging approach designed to capture the top and medial views of the foot simultaneously to allow rapid quantification of foot shape while eliminating exposure to ionizing radiation. We validated this system by longitudinally monitoring foot shape in a control group of 18 nonpregnant participants (mean ± SD age of 28 ± 4 years, mean ± SD BMI of 62.7 ± 6.4 kg) assessed 12 weeks apart. This control group included female patients who had no previous pregnancies, were older than 18 years of age, had a BMI less than 45, had no history of connective tissue disorders or trauma to the lower extremities that resulted in medical intervention, and could provide written consent. We collected PROMIS Pain Interference and Physical Function score assessments during pregnancy and postpartum. Both scores are standardized T-scores and range from 20 to 80, with higher Pain Interference and lower Physical Function scores indicating worse outcomes. Minimum clinically important differences for pain interference and physical function in foot and ankle orthopaedics range between 5.5 and 9.8 T-score points; therefore, we used a conservative threshold of > 10, which represents 1 SD, to identify clinically important changes in pain interference and physical function during pregnancy. Results: We found that foot shape changed from early to late pregnancy, with median (range) increases in left foot width (9.14 cm [8.42 to 11.50] versus 9.30 cm [8.55 to 10.57], median difference +0.21 cm [95% CI 0.06 to 0.25]; p = 0.02), left dorsum height (6.22 cm [5.52 to 7.12] versus 6.33 cm [5.67 to 7.78], median difference +0.17 cm [95% CI 0.01 to 0.46]; p = 0.03), and right dorsum height (6.30 cm [5.51 to 7.58] versus 6.50 cm [5.52 to 7.64], median difference +0.23 cm [95% CI 0.13 to 0.38]; p = 0.008). There were mean ± SD increases in left foot length (24.37 ± 1.24 cm versus 24.66 ± 1.12 cm, mean difference +0.29 cm [95% CI 0.09 to 0.49]; p = 0.03), left medial foot area (83.08 ± 7.12 cm 2 versus 87.30 ± 7.15 cm 2 , mean difference +4.22 cm 2 [95% CI 2.19 to 6.25]; p = 0.001), and right medial foot area (84.02 ± 7.71 cm 2 versus 88.10 ± 8.53 cm 2 , mean difference +4.08 cm 2 [95% CI 2.21 to 5.94]; p < 0.001). In addition, median (range) left foot width increased from early pregnancy to the postpartum period (9.14 cm [8.42 to 11.50] versus 9.42 cm [8.29 to 10.77], median difference +0.10 cm [95% CI 0.09 to 0.37]; p = 0.01). We found associations between foot shape and patient-reported pain during pregnancy. Pregnant participants with clinically important increases in pain interference from early to late pregnancy had increased right medial foot area during late pregnancy (median [range] 11.37 cm 2 /shoe size [9.72 to 15.97], median difference +1.02 cm 2 /shoe size [95% CI 0.42 to 1.73]; p = 0.005) and postpartum (11.26 cm 2 /shoe size [9.88 to 15.47], median difference +0.91 cm 2 /shoe size [95% CI 0.04 to 2.57]; p = 0.01) compared to the control group of nonpregnant participants (10.35 cm 2 /shoe size [8.76 to 11.54]). Pregnant participants with clinically important increases in pain interference from early to late pregnancy also demonstrated an increase in right foot length from early to late pregnancy when compared with nonpregnant participants who had no changes in pain interference (0.49 ± 0.59 cm versus -0.14 ± 0.49 cm, mean difference 0.63 cm [95% CI 0.21 to 1.06]; p = 0.005). Pregnant participants with clinically important declines in physical function had a greater increase in right medial foot area from early to late pregnancy than pregnant participants without a physical function decline (6.62 ± 4.94 cm 2 versus 1.98 ± 4.00 cm 2 , mean difference 4.64 cm 2 [95% CI 1.14 to 4.64]; p = 0.01). Conclusion: This study identified potential morphological risk factors for musculoskeletal pain and reduced function in pregnant patients. Patient-reported pain and function were predominantly associated with pregnancy-induced lower extremity edema. These findings have potential clinical impact to guide interventions (such as compression socks, massage, or foot elevation) aimed at managing foot changes during pregnancy to reduce musculoskeletal pain and disability. Level of Evidence Level III, prognostic study.
- Open Access
Two weeks of cumulative tendon load monitored by insole sensors is associated with plantar flexor function in Achilles tendinopathy
Tendon loading dictates rehabilitation outcomes in Achilles tendinopathy but is difficult to track in the real world. In this study, we used instrumented insole sensors to monitor Achilles tendon load for two weeks in fifteen individuals with Achilles tendinopathy, who also completed assessments of their plantar flexor strength, dynamic function, and survey-based outcomes. We used insole data to estimate two types of cumulative Achilles tendon load: overall (≥0.3×body weight) and high-level load (≥3×body weight). We determined Pearson correlations between (1) overall and high-level tendon loads, (2) plantar flexor moment, power, work, (3) heel raise height, repetitions, countermovement jump height, and (4) self-reported symptoms and activity. Overall cumulative tendon load moderately correlated to isometric plantar flexor moment (r = 0.543) and weakly to isokinetic and dynamic functions (0.128–0.413). Cumulative high-level tendon load strongly correlated to heel raise height (0.687) and fast isokinetic moment (0.625), and moderately to other functional measures (0.470–0.592). Symptoms weakly correlated to overall (0.392) and moderately to high-level load (0.436). Self-reported activity weakly correlated to overall (0.297) and strongly to high-level load (0.617). Stronger associations with the high-level Achilles tendon load than the overall load suggest that clinical function assessments provide insight into the real-world performance of high-loading activities. In contrast, the disconnect between overall tendon loading and plantar flexor function may explain the variability in recovery outcomes. Self-reported activity and standard heel raises represent high-level tendon load well, yet they do not always suggest functional deficit. Sensor-monitored tendon load shows promise as a new biomarker for real-world plantar flexor function in Achilles tendinopathy.
Open AccessTwo weeks of cumulative tendon load monitored by insole sensors is associated with plantar flexor function in Achilles tendinopathy
Tendon loading dictates rehabilitation outcomes in Achilles tendinopathy but is difficult to track in the real world. In this study, we used instrumented insole sensors to monitor Achilles tendon load for two weeks in fifteen individuals with Achilles tendinopathy, who also completed assessments of their plantar flexor strength, dynamic function, and survey-based outcomes. We used insole data to estimate two types of cumulative Achilles tendon load: overall (≥0.3×body weight) and high-level load (≥3×body weight). We determined Pearson correlations between (1) overall and high-level tendon loads, (2) plantar flexor moment, power, work, (3) heel raise height, repetitions, countermovement jump height, and (4) self-reported symptoms and activity. Overall cumulative tendon load moderately correlated to isometric plantar flexor moment (r = 0.543) and weakly to isokinetic and dynamic functions (0.128–0.413). Cumulative high-level tendon load strongly correlated to heel raise height (0.687) and fast isokinetic moment (0.625), and moderately to other functional measures (0.470–0.592). Symptoms weakly correlated to overall (0.392) and moderately to high-level load (0.436). Self-reported activity weakly correlated to overall (0.297) and strongly to high-level load (0.617). Stronger associations with the high-level Achilles tendon load than the overall load suggest that clinical function assessments provide insight into the real-world performance of high-loading activities. In contrast, the disconnect between overall tendon loading and plantar flexor function may explain the variability in recovery outcomes. Self-reported activity and standard heel raises represent high-level tendon load well, yet they do not always suggest functional deficit. Sensor-monitored tendon load shows promise as a new biomarker for real-world plantar flexor function in Achilles tendinopathy.
- Open Access
Wearable sensor and machine learning accurately estimate tendon load and walking speed during immobilizing boot ambulation
Achilles tendon injuries are treated with progressive weight bearing to promote tendon healing and restore function. Patient rehabilitation progression are typically studied in controlled, lab settings and do not represent the long-term loading experienced during daily living. The purpose of this study is to develop a wearable paradigm to accurately monitor Achilles tendon loading and walking speed using low-cost sensors that reduce subject burden. Ten healthy adults walked in an immobilizing boot under various heel wedge conditions (30°, 5°, 0°) and walking speeds. Three-dimensional motion capture, ground reaction force, and 6-axis inertial measurement unit (IMU) signals were collected per trial. We used Least Absolute Shrinkage and Selection Operator (LASSO) regression to predict peak Achilles tendon load and walking speed. The effects of using only accelerometer data, different sampling frequency, and multiple sensors to train the model were also explored. Walking speed models outperformed (mean absolute percentage error (MAPE): 8.41 ± 4.08%) tendon load models (MAPE: 33.93 ± 23.9%). Models trained with subject-specific data performed significantly better than generalized models. For example, our personalized model that was trained with only subject-specific data predicted tendon load with a 11.5 ± 4.41% MAPE and walking speed with a 4.50 ± 0.91% MAPE. Removing gyroscope channels, decreasing sampling frequency, and using combinations of sensors had inconsequential effects on models performance (changes in MAPE < 6.09%). We developed a simple monitoring paradigm that uses LASSO regression and wearable sensors to accurately predict Achilles tendon loading and walking speed while ambulating in an immobilizing boot. This paradigm provides a clinically implementable strategy to longitudinally monitor patient loading and activity while recovering from Achilles tendon injuries.
Open AccessWearable sensor and machine learning accurately estimate tendon load and walking speed during immobilizing boot ambulation
Achilles tendon injuries are treated with progressive weight bearing to promote tendon healing and restore function. Patient rehabilitation progression are typically studied in controlled, lab settings and do not represent the long-term loading experienced during daily living. The purpose of this study is to develop a wearable paradigm to accurately monitor Achilles tendon loading and walking speed using low-cost sensors that reduce subject burden. Ten healthy adults walked in an immobilizing boot under various heel wedge conditions (30°, 5°, 0°) and walking speeds. Three-dimensional motion capture, ground reaction force, and 6-axis inertial measurement unit (IMU) signals were collected per trial. We used Least Absolute Shrinkage and Selection Operator (LASSO) regression to predict peak Achilles tendon load and walking speed. The effects of using only accelerometer data, different sampling frequency, and multiple sensors to train the model were also explored. Walking speed models outperformed (mean absolute percentage error (MAPE): 8.41 ± 4.08%) tendon load models (MAPE: 33.93 ± 23.9%). Models trained with subject-specific data performed significantly better than generalized models. For example, our personalized model that was trained with only subject-specific data predicted tendon load with a 11.5 ± 4.41% MAPE and walking speed with a 4.50 ± 0.91% MAPE. Removing gyroscope channels, decreasing sampling frequency, and using combinations of sensors had inconsequential effects on models performance (changes in MAPE < 6.09%). We developed a simple monitoring paradigm that uses LASSO regression and wearable sensors to accurately predict Achilles tendon loading and walking speed while ambulating in an immobilizing boot. This paradigm provides a clinically implementable strategy to longitudinally monitor patient loading and activity while recovering from Achilles tendon injuries.