Christian Skalka
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Interests
Christian Skalka has a total of 27 co-authors. Here are the top co-authors by number of shared publications:
- Matthew Price
University of Vermont, Burlington, VT, USA
3 shared publications - Blake Hewgill
Department of Electrical and Biomedical Engineering, University of Vermont, Burlington, VT, United States
2 shared publications - Bruce D. Beynnon
Department of Orthopedics and Rehabilitation, McClure Musculoskeletal Research Center, Robert Larner College of Medicine, University of Vermont, Burlington, Vermont
2 shared publications - Jean Harvey
Department of Nutrition and Food Sciences, University of Vermont , Burlington, VT , USA
2 shared publications - Joseph P. Near
University of Vermont
2 shared publications - Michael J. Toth
University of Vermont College of Medicine
2 shared publications
Latest Publications
- 22 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 12 (24) , 52675-52686
Embedded IoT System for Acoustic Precipitation Phase Partitioning via Edge ML and MFCCs
Accurate and real-time detection of precipitation phases is essential for hydrological modeling, water resource management, and climate impact assessments. However, conventional methods struggle to distinguish precipitation phase near freezing and are unsuitable for distributed deployment in remote, complex terrain due to cost, size, and power constraints. In this work, we present an integrated acoustic sensing system that combines machine learning (ML) on the edge with acoustic sensing and Mel-frequency cepstral coefficients (MFCCs)-based feature extraction, all implemented on our designed edge device. The device is low-cost, with LAN and WAN capabilities for near real-time reporting and can integrate multiple sensors for snow hydrology studies, in addition to detecting the precipitation phase. We present the design, fabrication, and validation of the Advanced Unified Rainfall and Atmospheric Monitoring (AURA) system, integrating sensors for temperature, humidity, ultrasonic snow depth, solar radiation, and wind velocity. AURA features an embedded ML algorithm that enables accurate classification of precipitation phases. Short-time Fourier Transform (STFT) and MFCCs are computed on the edge from recorded precipitation acoustics via novel Mel filter bank computation methods. Using support vector machine (SVM) and random forest (RF) classifiers, the RF model achieves testing accuracy of 97.35% on simulated precipitation acoustics and 85.67% in consolidated class (CC) environmental recordings. The SVM classifier achieves 98.07% accuracy on simulated acoustics and 85.99% on CC environmental recordings. The low-power AURA network uses a long range (LoRa) star topology with time-division multiple access (TDMA) and an Iridium gateway for reliable data transfer from remote sites. Signal-to-noise ratio (SNR) analysis and comprehensive field tests confirm and validate system performance.
- 22 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 12 (24) , 52675-52686
Embedded IoT System for Acoustic Precipitation Phase Partitioning via Edge ML and MFCCs
Accurate and real-time detection of precipitation phases is essential for hydrological modeling, water resource management, and climate impact assessments. However, conventional methods struggle to distinguish precipitation phase near freezing and are unsuitable for distributed deployment in remote, complex terrain due to cost, size, and power constraints. In this work, we present an integrated acoustic sensing system that combines machine learning (ML) on the edge with acoustic sensing and Mel-frequency cepstral coefficients (MFCCs)-based feature extraction, all implemented on our designed edge device. The device is low-cost, with LAN and WAN capabilities for near real-time reporting and can integrate multiple sensors for snow hydrology studies, in addition to detecting the precipitation phase. We present the design, fabrication, and validation of the Advanced Unified Rainfall and Atmospheric Monitoring (AURA) system, integrating sensors for temperature, humidity, ultrasonic snow depth, solar radiation, and wind velocity. AURA features an embedded ML algorithm that enables accurate classification of precipitation phases. Short-time Fourier Transform (STFT) and MFCCs are computed on the edge from recorded precipitation acoustics via novel Mel filter bank computation methods. Using support vector machine (SVM) and random forest (RF) classifiers, the RF model achieves testing accuracy of 97.35% on simulated precipitation acoustics and 85.67% in consolidated class (CC) environmental recordings. The SVM classifier achieves 98.07% accuracy on simulated acoustics and 85.99% on CC environmental recordings. The low-power AURA network uses a long range (LoRa) star topology with time-division multiple access (TDMA) and an Iridium gateway for reliable data transfer from remote sites. Signal-to-noise ratio (SNR) analysis and comprehensive field tests confirm and validate system performance.
- Open Access
- 9 September 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-14
Language-Based Security for Low-Level MPC
Secure Multi-Party Computation (MPC) is an important enabling technology for data privacy in modern distributed applications. Currently, proof methods for low-level MPC protocols are primarily manual and thus tedious and error-prone, and are also non-standardized and unfamiliar to most PL theorists. As a step towards better language support and language-based enforcement, we develop a new staged PL for defining a variety of low-level probabilistic MPC protocols. We also formulate a collection of confidentiality and integrity hyperproperties for our language model that are familiar from information flow, including conditional noninterference, gradual release, and robust declassification. We demonstrate their relation to standard MPC threat models of passive and malicious security, and how they can be leveraged in security verification of protocols. To prove these properties we develop automated tactics in Math 1 that can be integrated with separation logic-style reasoning.
- 9 September 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-14
Open AccessLanguage-Based Security for Low-Level MPC
Secure Multi-Party Computation (MPC) is an important enabling technology for data privacy in modern distributed applications. Currently, proof methods for low-level MPC protocols are primarily manual and thus tedious and error-prone, and are also non-standardized and unfamiliar to most PL theorists. As a step towards better language support and language-based enforcement, we develop a new staged PL for defining a variety of low-level probabilistic MPC protocols. We also formulate a collection of confidentiality and integrity hyperproperties for our language model that are familiar from information flow, including conditional noninterference, gradual release, and robust declassification. We demonstrate their relation to standard MPC threat models of passive and malicious security, and how they can be leveraged in security verification of protocols. To prove these properties we develop automated tactics in Math 1 that can be integrated with separation logic-style reasoning.
- Open Access
- 15 October 2021
- Published by Association for Computing Machinery (ACM) in Digital Threats: Research and Practice
- Vol. 2 (4) , 1-29
Methods for Host-based Intrusion Detection with Deep Learning
Host-based Intrusion Detection Systems (HIDS) automatically detect events that indicate compromise by adversarial applications. HIDS are generally formulated as analyses of sequences of system events such as bash commands or system calls. Anomaly-based approaches to HIDS leverage models of normal (a.k.a. baseline) system behavior to detect and report abnormal events and have the advantage of being able to detect novel attacks. In this article, we develop a new method for anomaly-based HIDS using deep learning predictions of sequence-to-sequence behavior in system calls. Our proposed method, called the ALAD algorithm, aggregates predictions at the application level to detect anomalies. We investigate the use of several deep learning architectures, including WaveNet and several recurrent networks. We show that ALAD empowered with deep learning significantly outperforms previous approaches. We train and evaluate our models using an existing dataset, ADFA-LD, and a new dataset of our own construction, PLAID. As deep learning models are black box in nature, we use an alternate approach, allotaxonographs, to characterize and understand differences in baseline vs. attack sequences in HIDS datasets such as PLAID.
- 15 October 2021
- Published by Association for Computing Machinery (ACM) in Digital Threats: Research and Practice
- Vol. 2 (4) , 1-29
Open AccessMethods for Host-based Intrusion Detection with Deep Learning
Host-based Intrusion Detection Systems (HIDS) automatically detect events that indicate compromise by adversarial applications. HIDS are generally formulated as analyses of sequences of system events such as bash commands or system calls. Anomaly-based approaches to HIDS leverage models of normal (a.k.a. baseline) system behavior to detect and report abnormal events and have the advantage of being able to detect novel attacks. In this article, we develop a new method for anomaly-based HIDS using deep learning predictions of sequence-to-sequence behavior in system calls. Our proposed method, called the ALAD algorithm, aggregates predictions at the application level to detect anomalies. We investigate the use of several deep learning architectures, including WaveNet and several recurrent networks. We show that ALAD empowered with deep learning significantly outperforms previous approaches. We train and evaluate our models using an existing dataset, ADFA-LD, and a new dataset of our own construction, PLAID. As deep learning models are black box in nature, we use an alternate approach, allotaxonographs, to characterize and understand differences in baseline vs. attack sequences in HIDS datasets such as PLAID.
- Open Access
- 11 May 2020
- Published by JMIR Publications Inc. in JMIR Human Factors
- Vol. 7 (2) , e16605
A Cyber-Physical System for Near Real-Time Monitoring of At-Home Orthopedic Rehabilitation and Mobile–Based Provider-Patient Communications to Improve Adherence: Development and Formative Evaluation
Background: Knee extensor muscle performance is reduced after lower extremity trauma and orthopedic surgical interventions. At-home use of neuromuscular electrical stimulation (NMES) may improve functional recovery, but adherence to at-home interventions is low. Greater benefits from NMES may be realized with closer monitoring of adherence to at-home prescriptions and more frequent patient-provider interactions. Objective: This study aimed to develop a cyber-physical system to monitor at-home adherence to NMES prescription and facilitate patient-provider communications to improve adherence in near real time. Methods: The RehabTracker cyber-physical system was developed to accomplish this goal and comprises four components: (1) hardware modifications to a commercially available NMES therapy device to monitor device use and provide Bluetooth functionality; (2) an iPhone Operating System–based mobile health (mHealth) app that enables patient-provider communications in near real time; (3) a clinician portal to allow oversight of patient adherence with device use; and (4) a back-end server to store data, enable adherence analysis, and send automated push notifications to the patient. These four elements were designed to be fully compliant with the Health Insurance Portability and Accountability Act. The system underwent formative testing in a cohort of patients following anterior cruciate ligament rupture (n=7) to begin to assess face validity. Results: Compared with the NMES device software–tracked device use, the RehabTracker system recorded 83% (40/48) of the rehabilitation sessions, with 100% (32/32) of all sessions logged by the system in 4 out of 7 patients. In patients for whom tracking of automated push notifications was enabled, 100% (29/29) of the push notifications sent by the back-end server were received by the patient. Process, hardware, and software issues contributing to these inaccuracies are detailed. Conclusions: RehabTracker represents a promising mHealth app for tracking and improving adherence with at-home NMES rehabilitation programs and warrants further refinement and testing.
- 11 May 2020
- Published by JMIR Publications Inc. in JMIR Human Factors
- Vol. 7 (2) , e16605
Open AccessA Cyber-Physical System for Near Real-Time Monitoring of At-Home Orthopedic Rehabilitation and Mobile–Based Provider-Patient Communications to Improve Adherence: Development and Formative Evaluation
Background: Knee extensor muscle performance is reduced after lower extremity trauma and orthopedic surgical interventions. At-home use of neuromuscular electrical stimulation (NMES) may improve functional recovery, but adherence to at-home interventions is low. Greater benefits from NMES may be realized with closer monitoring of adherence to at-home prescriptions and more frequent patient-provider interactions. Objective: This study aimed to develop a cyber-physical system to monitor at-home adherence to NMES prescription and facilitate patient-provider communications to improve adherence in near real time. Methods: The RehabTracker cyber-physical system was developed to accomplish this goal and comprises four components: (1) hardware modifications to a commercially available NMES therapy device to monitor device use and provide Bluetooth functionality; (2) an iPhone Operating System–based mobile health (mHealth) app that enables patient-provider communications in near real time; (3) a clinician portal to allow oversight of patient adherence with device use; and (4) a back-end server to store data, enable adherence analysis, and send automated push notifications to the patient. These four elements were designed to be fully compliant with the Health Insurance Portability and Accountability Act. The system underwent formative testing in a cohort of patients following anterior cruciate ligament rupture (n=7) to begin to assess face validity. Results: Compared with the NMES device software–tracked device use, the RehabTracker system recorded 83% (40/48) of the rehabilitation sessions, with 100% (32/32) of all sessions logged by the system in 4 out of 7 patients. In patients for whom tracking of automated push notifications was enabled, 100% (29/29) of the push notifications sent by the back-end server were received by the patient. Process, hardware, and software issues contributing to these inaccuracies are detailed. Conclusions: RehabTracker represents a promising mHealth app for tracking and improving adherence with at-home NMES rehabilitation programs and warrants further refinement and testing.
- Open Access
- 8 October 2019
- Published by JMIR Publications Inc. in JMIR Preprints
A Cyber-Physical System for Near Real-Time Monitoring of At-Home Orthopedic Rehabilitation and Mobile–Based Provider-Patient Communications to Improve Adherence: Development and Formative Evaluation (Preprint)
BACKGROUND Knee extensor muscle performance is reduced after lower extremity trauma and orthopedic surgical interventions. At-home use of neuromuscular electrical stimulation (NMES) may improve functional recovery, but adherence to at-home interventions is low. Greater benefits from NMES may be realized with closer monitoring of adherence to at-home prescriptions and more frequent patient-provider interactions. OBJECTIVE This study aimed to develop a cyber-physical system to monitor at-home adherence to NMES prescription and facilitate patient-provider communications to improve adherence in near real time. METHODS The RehabTracker cyber-physical system was developed to accomplish this goal and comprises four components: (1) hardware modifications to a commercially available NMES therapy device to monitor device use and provide Bluetooth functionality; (2) an iPhone Operating System–based mobile health (mHealth) app that enables patient-provider communications in near real time; (3) a clinician portal to allow oversight of patient adherence with device use; and (4) a back-end server to store data, enable adherence analysis, and send automated push notifications to the patient. These four elements were designed to be fully compliant with the Health Insurance Portability and Accountability Act. The system underwent formative testing in a cohort of patients following anterior cruciate ligament rupture (n=7) to begin to assess face validity. RESULTS Compared with the NMES device software–tracked device use, the RehabTracker system recorded 83% (40/48) of the rehabilitation sessions, with 100% (32/32) of all sessions logged by the system in 4 out of 7 patients. In patients for whom tracking of automated push notifications was enabled, 100% (29/29) of the push notifications sent by the back-end server were received by the patient. Process, hardware, and software issues contributing to these inaccuracies are detailed. CONCLUSIONS RehabTracker represents a promising mHealth app for tracking and improving adherence with at-home NMES rehabilitation programs and warrants further refinement and testing. CLINICALTRIAL
- 8 October 2019
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessA Cyber-Physical System for Near Real-Time Monitoring of At-Home Orthopedic Rehabilitation and Mobile–Based Provider-Patient Communications to Improve Adherence: Development and Formative Evaluation (Preprint)
BACKGROUND Knee extensor muscle performance is reduced after lower extremity trauma and orthopedic surgical interventions. At-home use of neuromuscular electrical stimulation (NMES) may improve functional recovery, but adherence to at-home interventions is low. Greater benefits from NMES may be realized with closer monitoring of adherence to at-home prescriptions and more frequent patient-provider interactions. OBJECTIVE This study aimed to develop a cyber-physical system to monitor at-home adherence to NMES prescription and facilitate patient-provider communications to improve adherence in near real time. METHODS The RehabTracker cyber-physical system was developed to accomplish this goal and comprises four components: (1) hardware modifications to a commercially available NMES therapy device to monitor device use and provide Bluetooth functionality; (2) an iPhone Operating System–based mobile health (mHealth) app that enables patient-provider communications in near real time; (3) a clinician portal to allow oversight of patient adherence with device use; and (4) a back-end server to store data, enable adherence analysis, and send automated push notifications to the patient. These four elements were designed to be fully compliant with the Health Insurance Portability and Accountability Act. The system underwent formative testing in a cohort of patients following anterior cruciate ligament rupture (n=7) to begin to assess face validity. RESULTS Compared with the NMES device software–tracked device use, the RehabTracker system recorded 83% (40/48) of the rehabilitation sessions, with 100% (32/32) of all sessions logged by the system in 4 out of 7 patients. In patients for whom tracking of automated push notifications was enabled, 100% (29/29) of the push notifications sent by the back-end server were received by the patient. Process, hardware, and software issues contributing to these inaccuracies are detailed. CONCLUSIONS RehabTracker represents a promising mHealth app for tracking and improving adherence with at-home NMES rehabilitation programs and warrants further refinement and testing. CLINICALTRIAL
- Open Access
- 22 July 2019
- Published by JMIR Publications Inc. in JMIR Mental Health
- Vol. 6 (7) , e13946
Predicting Posttraumatic Stress Disorder Risk: A Machine Learning Approach
Background: A majority of adults in the United States are exposed to a potentially traumatic event but only a handful go on to develop impairing mental health conditions such as posttraumatic stress disorder (PTSD). Objective: Identifying those at elevated risk shortly after trauma exposure is a clinical challenge. The aim of this study was to develop computational methods to more effectively identify at-risk patients and, thereby, support better early interventions. Methods: We proposed machine learning (ML) induction of models to automatically predict elevated PTSD symptoms in patients 1 month after a trauma, using self-reported symptoms from data collected via smartphones. Results: We show that an ensemble model accurately predicts elevated PTSD symptoms, with an area under the curve (AUC) of .85, using a bag of support vector machines, naive Bayes, logistic regression, and random forest algorithms. Furthermore, we show that only 7 self-reported items (features) are needed to obtain this AUC. Most importantly, we show that accurate predictions can be made 10 to 20 days posttrauma. Conclusions: These results suggest that simple smartphone-based patient surveys, coupled with automated analysis using ML-trained models, can identify those at risk for developing elevated PTSD symptoms and thus target them for early intervention.
- 22 July 2019
- Published by JMIR Publications Inc. in JMIR Mental Health
- Vol. 6 (7) , e13946
Open AccessPredicting Posttraumatic Stress Disorder Risk: A Machine Learning Approach
Background: A majority of adults in the United States are exposed to a potentially traumatic event but only a handful go on to develop impairing mental health conditions such as posttraumatic stress disorder (PTSD). Objective: Identifying those at elevated risk shortly after trauma exposure is a clinical challenge. The aim of this study was to develop computational methods to more effectively identify at-risk patients and, thereby, support better early interventions. Methods: We proposed machine learning (ML) induction of models to automatically predict elevated PTSD symptoms in patients 1 month after a trauma, using self-reported symptoms from data collected via smartphones. Results: We show that an ensemble model accurately predicts elevated PTSD symptoms, with an area under the curve (AUC) of .85, using a bag of support vector machines, naive Bayes, logistic regression, and random forest algorithms. Furthermore, we show that only 7 self-reported items (features) are needed to obtain this AUC. Most importantly, we show that accurate predictions can be made 10 to 20 days posttrauma. Conclusions: These results suggest that simple smartphone-based patient surveys, coupled with automated analysis using ML-trained models, can identify those at risk for developing elevated PTSD symptoms and thus target them for early intervention.
- Open Access
- 8 March 2019
- Published by JMIR Publications Inc. in JMIR Preprints
Predicting Posttraumatic Stress Disorder Risk: A Machine Learning Approach (Preprint)
BACKGROUND A majority of adults in the United States are exposed to a potentially traumatic event but only a handful go on to develop impairing mental health conditions such as posttraumatic stress disorder (PTSD). OBJECTIVE Identifying those at elevated risk shortly after trauma exposure is a clinical challenge. The aim of this study was to develop computational methods to more effectively identify at-risk patients and, thereby, support better early interventions. METHODS We proposed machine learning (ML) induction of models to automatically predict elevated PTSD symptoms in patients 1 month after a trauma, using self-reported symptoms from data collected via smartphones. RESULTS We show that an ensemble model accurately predicts elevated PTSD symptoms, with an area under the curve (AUC) of .85, using a bag of support vector machines, naive Bayes, logistic regression, and random forest algorithms. Furthermore, we show that only 7 self-reported items (features) are needed to obtain this AUC. Most importantly, we show that accurate predictions can be made 10 to 20 days posttrauma. CONCLUSIONS These results suggest that simple smartphone-based patient surveys, coupled with automated analysis using ML-trained models, can identify those at risk for developing elevated PTSD symptoms and thus target them for early intervention.
- 8 March 2019
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessPredicting Posttraumatic Stress Disorder Risk: A Machine Learning Approach (Preprint)
BACKGROUND A majority of adults in the United States are exposed to a potentially traumatic event but only a handful go on to develop impairing mental health conditions such as posttraumatic stress disorder (PTSD). OBJECTIVE Identifying those at elevated risk shortly after trauma exposure is a clinical challenge. The aim of this study was to develop computational methods to more effectively identify at-risk patients and, thereby, support better early interventions. METHODS We proposed machine learning (ML) induction of models to automatically predict elevated PTSD symptoms in patients 1 month after a trauma, using self-reported symptoms from data collected via smartphones. RESULTS We show that an ensemble model accurately predicts elevated PTSD symptoms, with an area under the curve (AUC) of .85, using a bag of support vector machines, naive Bayes, logistic regression, and random forest algorithms. Furthermore, we show that only 7 self-reported items (features) are needed to obtain this AUC. Most importantly, we show that accurate predictions can be made 10 to 20 days posttrauma. CONCLUSIONS These results suggest that simple smartphone-based patient surveys, coupled with automated analysis using ML-trained models, can identify those at risk for developing elevated PTSD symptoms and thus target them for early intervention.
- Open Access
- 11 January 2016
- Published by JMIR Publications Inc. in JMIR Mental Health
- Vol. 3 (1) , e3
Usability Evaluation of a Mobile Monitoring System to Assess Symptoms After a Traumatic Injury: A Mixed-Methods Study
Background: Victims of trauma are at high risk for mental health conditions such as posttraumatic stress disorder and depression. Regular assessment of mental health symptoms in the post-trauma period is necessary to identify those at greatest risk and provide treatment. The multiple demands of the acute post-trauma period present numerous barriers to such assessments. Mobile apps are a method by which to overcome these barriers in order to regularly assess symptoms, identify those at risk, and connect patients to needed services. Objective: The current study conducted a usability evaluation of a system to monitor mental health symptoms after a trauma. The system was developed to promote ease of use and facilitate quick transmission of data. Methods: A sample of 21 adults with a history of trauma completed a standardized usability test in a laboratory setting followed by a qualitative interview. Results: Usability testing indicated that the app was easy to use and that patients were able to answer several questions in less than 1 minute (mean [SD] 29.37 [7.53]; range 15-57). Qualitative analyses suggested that feedback should be included in such an app and recommendations for the type of feedback were offered. Conclusions: The results of the current study indicate that a mobile app to monitor post-trauma mental health symptoms would be well received by victims. Personalized feedback to the user was identified as critical to promote the usability of the software.
- 11 January 2016
- Published by JMIR Publications Inc. in JMIR Mental Health
- Vol. 3 (1) , e3
Open AccessUsability Evaluation of a Mobile Monitoring System to Assess Symptoms After a Traumatic Injury: A Mixed-Methods Study
Background: Victims of trauma are at high risk for mental health conditions such as posttraumatic stress disorder and depression. Regular assessment of mental health symptoms in the post-trauma period is necessary to identify those at greatest risk and provide treatment. The multiple demands of the acute post-trauma period present numerous barriers to such assessments. Mobile apps are a method by which to overcome these barriers in order to regularly assess symptoms, identify those at risk, and connect patients to needed services. Objective: The current study conducted a usability evaluation of a system to monitor mental health symptoms after a trauma. The system was developed to promote ease of use and facilitate quick transmission of data. Methods: A sample of 21 adults with a history of trauma completed a standardized usability test in a laboratory setting followed by a qualitative interview. Results: Usability testing indicated that the app was easy to use and that patients were able to answer several questions in less than 1 minute (mean [SD] 29.37 [7.53]; range 15-57). Qualitative analyses suggested that feedback should be included in such an app and recommendations for the type of feedback were offered. Conclusions: The results of the current study indicate that a mobile app to monitor post-trauma mental health symptoms would be well received by victims. Personalized feedback to the user was identified as critical to promote the usability of the software.