User Identification from Time Series of Fitness Data

Marchioro, Thomas; Kazlouski, Andrei; Markatos, Evangelos P.

doi:10.5220/0010585008060811

Cited by 3 publications

(3 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Previous methods focusing on person recognition using gesture data have been reported. 27 , 28 In contrast to our study, previous signature identification research primarily investigated security and safety to avoid unintended intrusions. For instance, 1 group developed RFnet, that is, a multi-branch 1D-CNN network, for classifying gestures using time-series data.…”

Section: Discussionmentioning

confidence: 91%

“… 28 Health and fitness data collected from wearable devices has been studied as a unique identifier of a person, which can pose a potential threat to targeted advertisement and violation of privacy rights. 27 Additionally, current literature on using deep learning methods in surgery sought to characterize surgeon skill level and surgical task identification using one or a combination of video object tracking or detection, spatiotemporal video descriptors, robotic kinematics, and virtual reality interfaces. 29 – 35 Furthermore, limitations of previous studies include the use of the da Vinci Surgical System on bench-top surgical training models rather than direct manipulation of the surgical tools during clinical cases.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Force Profile as Surgeon-Specific Signature

Baghdadi,

Guo,

Lama

et al. 2023

Annals of Surgery Open

View full text Add to dashboard Cite

Objective: To investigate the notion that a surgeon’s force profile can be the signature of their identity and performance. Summary background data: Surgeon performance in the operating room is an understudied topic. The advent of deep learning methods paired with a sensorized surgical device presents an opportunity to incorporate quantitative insight into surgical performance and processes. Using a device called the SmartForceps System and through automated analytics, we have previously reported surgeon force profile, surgical skill, and task classification. However, an investigation of whether an individual surgeon can be identified by surgical technique has yet to be studied. Methods: In this study, we investigate multiple neural network architectures to identify the surgeon associated with their time-series tool-tissue forces using bipolar forceps data. The surgeon associated with each 10-second window of force data was labeled, and the data were randomly split into 80% for model training and validation (10% validation) and 20% for testing. Data imbalance was mitigated through subsampling from more populated classes with a random size adjustment based on 0.1% of sample counts in the respective class. An exploratory analysis of force segments was performed to investigate underlying patterns differentiating individual surgical techniques. Results: In a dataset of 2819 ten-second time segments from 89 neurosurgical cases, the best-performing model achieved a micro-average area under the curve of 0.97, a testing F1-score of 0.82, a sensitivity of 82%, and a precision of 82%. This model was a time-series ResNet model to extract features from the time-series data followed by a linearized output into the XGBoost algorithm. Furthermore, we found that convolutional neural networks outperformed long short-term memory networks in performance and speed. Using a weighted average approach, an ensemble model was able to identify an expert surgeon with 83.8% accuracy using a validation dataset. Conclusions: Our results demonstrate that each surgeon has a unique force profile amenable to identification using deep learning methods. We anticipate our models will enable a quantitative framework to provide bespoke feedback to surgeons and to track their skill progression longitudinally. Furthermore, the ability to recognize individual surgeons introduces the mechanism of correlating outcome to surgeon performance.

show abstract

Section: Discussionmentioning

confidence: 91%

Section: Discussionmentioning

confidence: 99%

Force Profile as Surgeon-Specific Signature

Baghdadi,

Guo,

Lama

et al. 2023

Annals of Surgery Open

View full text Add to dashboard Cite

show abstract

“…Activity logs themselves often contain information that is derived from an individual's attributes: calories, for instance, are often estimated by combining activity information with physical characteristics such as age, gender, height, and weight [5]. Thus, an ad-versary can leverage additional records that she possesses to find a target user in the dataset, even when personal attributes are removed [6], and glean insights about them. Such insights may include:…”

Section: Privacy Aspects Of Sharing Data From Wearablesmentioning

confidence: 99%

How to Publish Wearables’ Data: Practical Guidelines to Protect User Privacy

Marchioro

Kazlouski

Markatos

2022

Studies in Health Technology and Informatics

View full text Add to dashboard Cite

The spread of wearable fitness trackers has contributed to the increase of various fitness studies, utilizing such devices to collect data from a group of users. When these data are disclosed, the lack of sanitization can lead to severe privacy risks. In this paper, we discuss the various threats that are posed by disclosing unaltered information from wearable devices. We also dismiss common fallacies of fitness data sharing and present practical guidelines to preserve user privacy.

show abstract