Salim Chemlal

Blood glucose individualized prediction for type 2 diabetes using iPhone application

Type 2 diabetes is now the most rapidly growing form of diabetes and has become increasingly common among children. This paper presents our work of implementing an individualized real time predictive system for blood glucose in type 2 diabetes in an iPhone application. The developed application, called HealthiManage, provides relevant feedback to patients at each glucose input reading comparing the measured and predicted readings, facilitating improved self-management of the disease. The application incorporates activity recognition via a built-in accelerometer on the iPhone, which monitors any physical activity and adjusts predictions accordingly. Also, a reward component interface was incorporated that is intended to enhance patient compliance and encourage mainly teenagers to take control and improve their blood glucose regulation. The individualized prediction algorithm was tested and verified with real patient data. Different physical activities were also examined and classified for an accurate activity recognition component. The designed application with its predictive model, activity recognition, and other elements provide what we believe to be helpful feedback to monitor and manage type 2 diabetes and improve patient compliance

Orientation invariant ECG-based stethoscope tracking for heart auscultation training on augmented standardized patients

Auscultation, the act of listening to the heart and lung sounds, can reveal substantial information about patients’ health and other cardiac-related problems; therefore, competent training can be a key for accurate and reliable diagnosis. Standardized patients (SPs), who are healthy individuals trained to portray real patients, have been extensively used for such training and other medical teaching techniques; however, the range of symptoms and conditions they can simulate remains limited since they are only patient actors. In this work, we describe a novel tracking method for placing virtual symptoms in correct auscultation areas based on recorded ECG signals with various stethoscope diaphragm orientations; this augmented reality simulation would extend the capabilities of SPs and allow medical trainees to hear abnormal heart and lung sounds in a normal SP. ECG signals recorded from two different SPs over a wide range of stethoscope diaphragm orientations were processed and analyzed to accurately distinguish four different heart auscultation areas, aortic, mitral, pulmonic and tricuspid, for any stethoscope’s orientation. After processing the signals and extracting relevant features, different classifiers were applied for assessment of the proposed method; 95.1% and 87.1% accuracy were obtained for SP1 and SP2, respectively. The proposed system provides an efficient, non-invasive, and cost efficient method for training medical practitioners on heart auscultation.

Using EKG signals for virtual pathology stethoscope tracking in standardized patient heart auscultation

Standardized patients (SPs) have been widely used in medical education to improve the clinical skills of medical practitioners. SPs are trained to realistically portray real patients with specific illnesses and conditions. We have been using augmented standardized patients (ASPs) to extend the capabilities of SPs by utilizing medical instruments that can emplace virtual pathology to enhance the experience and variety of symptoms that medical students can encounter with the otherwise healthy SPs. In this paper, we describe a realistic tracking method for virtual pathology stethoscopes that provide an efficient and cost effective method for placing virtual symptoms in correct auscultation locations. The tracking method involves analyzing and processing EKG signals of the heart which can be implemented in augmented stethoscopes for the presentation of normal and abnormal cardiac diseases.

Simulation of the critical steps of the Nuss procedure

Simulation can be a critical component in the surgical training, planning and rehearsal process of difficult or new procedures. This is true for the Nuss procedure, which is a minimally invasive surgery for correcting pectus excavatum (PE) – a congenital chest wall deformity. Surgeons who routinely perform this surgery identified the most critical steps during the procedure. We chose to simulate these components which involve the process of creating a pathway anterior to the pericardium underneath the sternum as this step can, if incorrectly performed, lead to the death of a patient or, at least, serious complications. In this paper, we describe the hardware setup of the Nuss procedure surgical simulator, the constructed virtual environment, modelling of anatomical structures, surgical instrument and its insertion point without the aid of physical components, reproducing PE deformity, and eventually physics of the tunnelling process.

HealthiManage: An Individualized Prediction Algorithm for Type 2 Diabetes Chronic Disease Control

This paper describes a prediction algorithm for blood glucose in Type 2 diabetes. An iPhone application was developed that allows patients to record their daily blood glucose levels and provide them with relevant feedback using the prediction algorithm to help control their blood glucose levels. Several methods using theoretical functions were tested to select the most accurate prediction method. The prediction is adjusted with each glucose reading input by the patient taking into consideration the time of the glucose reading and the time after the patient’s last meal, as well as any physical activity. The individualized prediction algorithm was tested and verified with real patient data and also validated using a non-parametric regression method. The accuracy of prediction results varied from different approaches and was adequate for most of the methods tested. The predicted results merged closer to the patients’ actual glucose readings after each additional input reading. The findings of the research were encouraging and the predictive system provided what we believe to be helpful feedback to control, improve, and take proactive measures to regulate blood glucose levels.

Poster: Modeling insertion point for general purpose haptic device simulations for minimally invasive surgeries

Minimally invasive surgery has revolutionized surgical procedures over the last decades. This trend has impacted the surgical simulation field and promoted a wide use of haptic devices. The majority of minimally invasive surgery simulators follow a visuo-haptic setup where ports are mounted stationary and physical constraints are used to simulate the pivot of surgical tools. This setup may add limitations to certain procedures and result in negative training, as those assumptions may not hold true during surgery. In this work, we investigate challenges in modeling general purpose haptic device simulations for minimally invasive surgeries, namely accurate pivoting around the insertion point, collisions with surrounding objects during procedure, and system scalability for various devices. The system behavior is evaluated on a Nuss procedure surgical simulator and tested by experienced surgeons on the field. The findings are promising and may benefit other surgical simulators in which the kinematics and dynamics of the surgical tool are utilized within the context of minimally invasive surgeries.

Improvement of a Virtual Pivot for Minimally Invasive Surgery Simulators Using Haptic Augmentation

With rapid development of minimally invasive surgery, proficiency with intricate skills is becoming a greater concern. Consequently, the use of out-of-operating room training has increased significantly through employing high-fidelity and anatomically-correct graphics and haptic interfaces in virtual reality simulations. The effort in developing surgical simulators for generic minimally invasive procedures is still, however, suboptimal for many haptic implementations. A main aspect of such simulations is the pivoting behavior of the surgical tool realized using the haptic device. This paper investigates the limitation of a fully-virtual implementation of the pivot and the ability to augment haptic interfaces to achieve a natural representation of forces. The design and implementation of two surgical tool pivoting techniques are introduced. Furthermore, a phantom is constructed from synthesized components to be used to measure and reproduce realistic mechanical properties of the anatomical model and pivot behavior.