Search

BACKGROUND CONTEXT Traumatic spinal cord injury can have a dramatic effect on a patient's life. The degree of neurologic recovery greatly influences a patient's treatment and expected quality of life. This has resulted in the development of machine learning algorithms (MLA) that use acute demographic and neurologic information to prognosticate recovery. The van Middendorp et al. (2011) (vM) logistic regression (LR) model has been established as a reference model for the prediction of walking recovery following spinal cord injury as it has been validated within many different countries. However, an examination of the way in which these prediction models are evaluated is warranted. The area under the receiver operators curve (AUROC) has been consistently used when evaluating model performance, but it has been shown that AUROC overemphasizes the most common event resulting in an inaccurate assessment when the data are imbalanced. Furthermore, there is evidence that the use of more advanced MLA, such as an unsupervised k-means model, may show superior performance compared to LR as they can handle a larger number of features. PURPOSE The first objective of the study was to assess the performance of both an unsupervised MLA and LR model with complete admission neurologic information against the vM and Hicks models. Second, a comparison between the accuracy of the AUROC and the F1-score will be made to determine which method is superior for the assessment of diagnostic performance of prediction models on large-scale datasets. STUDY DESIGN Retrospective review of a prospective cohort study. PATIENT SAMPLE The Rick Hansen Spinal Cord Injury Registry (RHSCIR) was used in this study. All patients enrolled between 2004 and 2017 with complete neurologic examination and Functional Independence Measure outcome data at ≥1 year follow-up or who could walk at discharge were included. The prognostic variables included age (dichotomized at ≥65 years old); American Spinal Injury Association Impairment Scale (AIS) grade; and individual motor, light touch, and pinprick score from L2 to S1. OUTCOME MEASURES The Functional Independence Measure locomotor score was used to assess independent walking ability at discharge or 1-year follow-up. METHODS An unsupervised MLA with k=2 was chosen in order to identify a “walk” cluster and a “not walk” cluster. Model performance was assessed through the development of a receiver operating characteristic curve with associated AUROC and a precision-recall curve with associated F1-score. The study and the RHSCIR are supported by funding from Health Canada, Western Economic Diversification Canada, and the Governments of Alberta, British Columbia, Manitoba, and Ontario. These funders had no role in the study or study reporting and the authors have no conflicts of interest to report. RESULTS No clinically relevant differences were found between with the use of an unsupervised MLA with a greater amount of initial neurologic information compared to the established standards for any AIS classification. Although demonstrated for all separate AIS classifications, most notably, the AUROC for the vM (0.78) and Hicks models (0.76) were found to be superior to that of the new LR model (0.72); however, the vM and Hicks models had more than double the amount of false negative classifications compared to the LR. The F1-scores between these three models were also found to be different but with the vM and Hicks models being lower than the LR (0.85, 0.81, and 0.89, respectively). CONCLUSIONS No clinically relevant differences were found between the use of an unsupervised MLA with complete admission neurologic information compared to the previously validated standards; however, when comparing the performance of the AUROC and F1-score, the AUROC showed inaccurate prognostic performance when there was an imbalance toward a greater amount of false negatives. Importantly, the F1-score did not succumb to this imbalance. As AUROC has been used as the standard when evaluating performance of prediction models, consideration as to whether this is the most appropriate method is warranted. Future work should focus on comparing AUROC and F1-scores with other previously validated models.

Recently, the focus on sentiment analysis has been domain dependent even though the expressions used by the public are unsophisticatedly familiar regardless of the topics or domains. Online social media (OSNs) has been a daily venue for informal conversational contents from various domains ranging from sports and cooking to politics and human rights. Generating specific resources for every domain independently requires high cost and extensive efforts. In response, we propose to build a general multi-class sentiment classifier using our Domain-Free Sentiment Multimedia Dataset (DFSMD). Based on the proven capabilities of Light Gradient Boosting Machine (LGBM) in dealing with high dimensional and imbalance data, we have trained an LGBM model to recognize one of three sentiments of tweets: positive, negative, or neutral. We have conducted extensive comparisons and evaluations for six other standard sentiment classification algorithms and different sets of features including OSNs-specific ones. Our results have shown that LGBM model is the winner among the other six algorithms. It has been also shown that our dataset contains distinguishing characteristics in the three classes. Moreover, hashtag words are shown to be significantly important in capturing the sentiments of tweets. In addition, our findings have revealed the effectiveness of our approach in adapting general-domain sentiment to domain-specific sentiment analysis.

BACKGROUND CONTEXT Resident involvement in the operating room is a vital component of their medical education. Conflicting and limited research exists regarding the effects of surgical resident participation on spine surgery patient outcomes. PURPOSE To determine the effect of resident involvement on surgery duration, length of hospital stay and 30-day postoperative complication rates in common spinal surgery using the American College of Surgeons’ National Surgical Quality Improvement Program (ACS-NSQIP) database. STUDY DESIGN Multicenter retrospective cohort study. PATIENT SAMPLE A total of 1,441 patients met the inclusion criteria: 1,142 patients had surgeries with an attending physician alone and 299 patients had surgeries with trainee involvement. All anterior cervical or posterior lumbar surgery patients were identified. Patients who had missing trainee involvement information, surgery for cancer, preoperative infection or dirty wound classification, spine fractures, traumatic spinal cord injury, intradural surgery, thoracic surgery, and emergency surgery were excluded. OUTCOME MEASURES The main outcomes of interest analyzed from the ACS-NSQIP database included surgical complications, medical complications, length of hospital stay, and surgery duration. METHODS Propensity score for risk of any complication was calculated to account for baseline characteristic differences between the attending alone and trainee present group. Multivariate logistic regression was used to investigate the impact of resident involvement on surgery duration, length of hospital stay, and 30-day postoperative complication rates. RESULTS After adjusting using the calculated propensity score, the multivariate analysis demonstrated that there was no significant difference in any complication rates between surgeries involving trainees compared to surgeries with attending surgeons alone. Surgery times were found to be significantly longer for surgeries involving trainees. To further explore this relationship, separate analyses were performed for tertiles of predicted surgery duration, cervical or lumbar surgery, fusion or nonfusion, and inpatient or outpatient surgery. The effect of trainee involvement on increasing surgery time remained significant for medium predicted surgery duration, longer predicted surgery duration, cervical surgery, lumbar surgery, fusion surgery, and inpatient surgery. There were no significant differences reported for any other factors. CONCLUSIONS After adjusting for confounding, we demonstrated in a national database that resident involvement in surgeries did not increase complication rates. We demonstrated that surgeries with more complex features may lead to an increase in operative time when trainees are involved. Further study is required to determine how to efficiently integrate resident involvement in surgeries without affecting their medical education.

BACKGROUND With spinal surgery rates increasing in North America, models that are able to accurately predict which patients are at greater risk of developing complications are highly warranted. However, the previously published methods which have used large, multi-centre databases to develop their prediction models have relied on the receiver operator characteristics curve with the associated area under the curve (AUC) to assess their model's performance. Recently, it has been found that a precision-recall curve with the associated F1-score could provide a more realistic analysis for these models. PURPOSE To develop a logistic regression (LR) model for the prediction of complications following posterior lumbar spine surgery and to then assess for any difference in performance of the model when using the AUC versus the F1-score. STUDY DESIGN Retrospective review of a prospective cohort. PATIENT SAMPLE The American College of Surgeons National Surgical Quality Improvement Program (NSQIP) registry was used. All patients that underwent posterior lumbar spine surgery between 2005 to 2016 with appropriate data were included. OUTCOME MEASURES Both the AUC and F1-score were utilized to assess the prognostic performance of the prediction model. METHODS In order to develop the LR model used to predict a complication during or following spine surgery, 19 variables were selected by three orthopedic spine surgeons from the NSQIP registry. Two datasets were developed for this analysis: (1) an imbalanced dataset, which was taken directly from the NSQIP registry, and (2) a down-sampled set. The purpose of the down-sampled set was to balance the data in order to evaluate whether balancing the data had an effect on model performance. The AUC and F1-score were applied to both of these datasets. RESULTS Within the NSQIP database, 52,787 spine surgery cases were identified of which only 10% of these cases had complications during surgery. Applying the LR model showed a large difference between the AUC (0.69) and the F1 score (0.075) on the imbalanced dataset. However, no major differences existed between the AUC and F1-score when the data was balanced and the LR model was reapplied (0.69 and 0.62, AUC and F1-score, respectively). CONCLUSIONS The F1-score detected a drastically lower performance for the prediction of complications when using the imbalanced data, but detected a performance similar to the AUC level when balancing techniques were utilized for the dataset. This difference is due to a low precision score when many false positive classifications are present, which is not identified when using the AUC value. This lowers the utility of the AUC score, as many of the datasets used in medicine are imbalanced. Therefore, we recommend using the F1-score on large, prospective databases when the data is imbalanced with a large amount of true negative classifications.

Given the flexibility and potential of cloud technologies, cloud-based rehabilitation frameworks have shown encouraging results as assistive tools for post-stroke disability rehabilitation exercises and treatment. To treat post-stroke disability, cloud-based rehabilitation offers great advantages over conventional, clinic-based rehabilitation, providing ubiquitous flexible rehabilitation services and storage while offering therapeutic feedback from a therapist in real-time during patients' rehabilitative movements. With the development of sensory technologies, cloud computing technology integrated with Augmented Reality (AR) may make therapeutic exercises more enjoyable. To achieve these objectives, this paper proposes a framework for cloud-based rehabilitation services, which uses AR technology along with other sensory technologies. We have designed a prototype of the framework that uses the mechanism of sensor gloves to recognize gestures, detecting the real-time condition of a patient doing rehabilitative exercises. This prototype framework is tested on twelve patients not using sensor gloves and on four patients wearing sensor gloves over six weeks. We found statistically significant differences between the forces exerted by patients’ fingers at week one compared to week six. Significant improvements in finger strength were found after six weeks of therapeutic rehabilitative exercises.

It is well known that home exercise is as good as the rehabilitation center. Unfortunately, passive devices such as dumbbells, elastic bands, stress balls, and tubing, which have been widely used for home-based upper-body rehabilitation, do not provide therapists with the information needed to monitor patients’ progress, identify impairments, and suggest treatments. Moreover, the lack of interactivity of these devices turns rehabilitation exercises into boring, unpleasant tasks. In this chapter, we introduce a family of exergame rehabilitation systems aimed at solving the aforementioned problems. The systems combine recent rehabilitation approaches with efficient, yet affordable, skeleton tracking input technologies, and multimodal interactive computer environment. In addition, the systems provide real-time feedback to stroke patients, summarize feedback after each session, and predict the overall recovery progress. Moreover, these systems show a new style of rehabilitation that motivates patients by engaging family and friends in the rehabilitation process and allowing therapists to remotely assess the progress of patients and adjust the training strategy accordingly. The objective/subjective assessments and usability studies show the feasibility of the proposed systems for rehabilitation in stroke patients with upper limb motor dysfunction.

Predicting surgical complications can improve shared decision making by surgeons and patients. Recently, the use of machine learning algorithms for predicting complications has gained much attention. In this study, we used the American college of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) database to compare the performance of five machine learning algorithms for predicting complications during spine surgery. The database included 173449 patients who underwent spine surgery. To thoroughly evaluate and compare the proposed machine learning algorithms, the dataset was balanced and the algorithms were applied on both the balanced and imbalanced dataset. The results indicated that no significant difference was found between the AUCs for machine learning models of the imbalanced and balanced dataset. However, when the f1 score was considered as a metric, the performance of the machine learning models trained with the balanced dataset had significantly outperformed those algorithms trained with the imbalanced dataset.

Muscle strength is mostly measured by wearable devices. However, wearing such devices is a tedious, unpleasant, and sometimes impossible task for stroke patients. In this paper, a mathematical model is proposed to estimate the strength of the upper limb muscles of a stroke patient by using Microsoft Kinect sensor. A prototype exergame is designed and developed to mimic real post-stroke rehabilitation exercises. Least-square regression matrix is used to find the relation between the kinematics of the upper limb and the strength of the corresponding muscles. Kinect sensor is used along with a force sensing resistors (FSR) glove and two straps to collect both, real-time upper limb joints data and the strength of muscles of the subjects while they are performing the exercises. The prototype of this system is tested on five stroke patients and eight healthy subjects. Results show that there is no statistically significant difference between the measured and the estimated values of the upper-limb muscles of the stroke patients. Thus, the proposed method is useful in estimating the strength of the muscles of stroke patient without the need to wear any devices.

The proliferation of online soccer contents generated by fans and teams has been widely leveraged in the area of sentiment analysis research. This work proposes domain-specific approach for understanding sentiment expressed in soccer conversations. We introduce a soccer-specific lexicon that we leverage in building sentiment model trained on soccer dataset. Our results show the effectiveness of the proposed approach in recognizing fans emotion during soccer events.

Throughout the last decade, there has been a dramatic decline in daily physical activity among individuals which results in numerous severe health issues, including obesity, cardiovascular diseases, and high blood pressure. Pervasive applications that promote health living present a promising approach for the reduction or prevention of these health ailements. Context-awareness plays an important role in the development of these technologies. Context-aware systems can adapt to the changing context in order to better serve their users. In this paper, we propose the Extensible Context Ontology for Persuasive Physical-Activity Applications (ECOPPA), a formal context modeling scheme for applications that promote physical activity.

Nowadays, a great number of patients around the world require rehabilitation programs. Virtual and augmented reality has been introduced as promising assistive tools to support the patients, allowing them to continue their recovery training by themselves. The purpose of this study is to develop an intelligent, low- cost rehabilitation gaming system using a Microsoft Kinect sensor. We have developed two goal-oriented exercises that target the rehabilitation of patients with chronic pain of the arm. These exercises include moving a ball into a net (basketball game) in the vertical plane and reaching a cup (reaching game) in the horizontal plane. In addition, accurate performance measures are provided for each exercise to quantitatively evaluate the effectiveness of the treatment plan for each patient. The implemented system has increased user motivation to participate in the rehabilitation program, accelerated the recovery of the muscles, and offered the therapists the ability to assess the patients' progress.

In this paper, we propose a framework of a cost-effective, entertaining, and motivating home-based upper limb rehabilitation system which consists of a cloud system and a client interface. The framework provides real-time feedback to the patient subject, summarizes the feed-back after each session, and predicts the rehabilitation performance. As an implementation of the framework, a Kinect sensor is used to collect real-time data for upper limb joints of the subjects while they are participating in rehabilitation exergames. The Dynamic Time Warping DTW algorithm is then applied to compare the movement pattern of a patient subject with the movement pattern of a healthy subject. Next, the Auto-Regressive Integrated Moving Average ARIMA is utilized to forecast the rehabilitation progress of the patients based on their performance history. The prototype of this system is tested on six healthy individuals and one patient. The results show that the patients' movement patterns have a similar curve shape to the healthy individuals' movement patterns and, hence, the DTW algorithm can be used as an effective index to describe the rehabilitation statuses of the subjects. The forecasting method is briefly tested by feeding the rehabilitation status history.

Obesity has become a worldwide problem which most countries are trying to fight. It affects many people, irrespective of age, race, gender, or religion, anyone can suffer from obesity that leads to serious problems for individuals and for society as a whole. In this study we have selected two groups of people: the basic people who rarely exercise on a weekly basis, and the average people who exercise regularly every week. We have explored the attitude of the two groups towards mixing exercises with games in order to motivate the people with basic activity levels to exercise more frequently. We have used a qualitative standard online questionnaire from AttrakDiff and we have done a quantitative study of some important factors during exercises. The results of the qualitative and quantitative studies were very encouraging, as they reveal that mixing games with exercises can transform boring exercises into entertaining ones. It can also motivate players to continue and repeat the exercises. The ANOVA test has been applied and it shows that combining games with the bike has a significant effect on the speed and the average rotation per minute of the participants.