A New Hybrid Possibilistic-Probabilistic Decision-Making Scheme for Classification.

Uncertainty is at the heart of decision-making processes in most real-world applications. Uncertainty can be broadly categorized into two types: aleatory and epistemic. Aleatory uncertainty describes the variability in the physical system where sensors provide information (hard) of a probabilistic type. Epistemic uncertainty appears when the information is incomplete or vague such as judgments or human expert appreciations in linguistic form. Linguistic information (soft) typically introduces a possibilistic type of uncertainty. This paper is concerned with the problem of classification where the available information, concerning the observed features, may be of a probabilistic nature for some features, and of a possibilistic nature for some others. In this configuration, most encountered studies transform one of the two information types into the other form, and then apply either classical Bayesian-based or possibilistic-based decision-making criteria. In this paper, a new hybrid decision-making scheme is proposed for classification when hard and soft information sources are present. A new Possibilistic Maximum Likelihood (PML) criterion is introduced to improve classification rates compared to a classical approach using only information from hard sources. The proposed PML allows to jointly exploit both probabilistic and possibilistic sources within the same probabilistic decision-making framework, without imposing to convert the possibilistic sources into probabilistic ones, and vice versa.

Iterative Refinement of Possibility Distributions by Learning for Pixel-Based Classification.

This paper proposes an approach referred as: iterative refinement of possibility distributions by learning (IRPDL) for pixel-based image classification. The IRPDL approach is based on the use of possibilistic reasoning concepts exploiting expert knowledge sources as well as ground possibilistic seeds learning. The set of seeds is constructed by incrementally updating and refining the possibility distributions. Synthetic images as well as real images from the RIDER Breast MRI database are being used to evaluate the IRPDL performance. Its performance is compared with three relevant reference methods: region growing, semi-supervised fuzzy pattern matching, and Markov random fields. The IRDPL performance (in terms of recognition rate, 87.3%) is close to the Markovian method (88.8%) that is considered to be the reference in pixel-based image classification. IRPDL outperforms the other two methods, respectively, at the recognition rates of 83.9% and 84.7%. In addition, the proposed IRPDL requires fewer parameters for the mathematical representation and presents a reduced computational complexity.

Finite element lifetime prediction of a miniature adjustable orthopedic device.

In Total Knee Arthroplasty (TKA), accurate balancing of the medial and lateral collateral ligaments is considered by orthopedic surgeons as one of the most challenging and complicated tasks to achieve. Therefore, an efficient solution is needed to assist the surgeons in achieving this crucial task without resulting in tibiofemoral misalignment. The required solution consists in developing either a completely automated smart ligament balancer for intraoperative use or adjustable tibial implant for postoperative use. The smart ligament balancer allows the surgeon to accurately balance the collateral ligaments at the time of surgery while the adjustable tibial implant can be controlled in the postoperative period in order to correct the residual ligament imbalance. In this paper, we propose a miniature device that can be used as a smart ligament balancer during TKA or as an adjustable tibial implant in the period following the surgery. Three designs of the smart ligament balancer have been developed using 3-Dimensional (3D) Computer Assisted Design (CAD) software. The proposed balancer can also be used as an adjustable tibial implant after slightly modifying its design. Finite element study of each design has been conducted in order to predict the lifetime of this implant in both cases of intraoperative or postoperative uses.

A self-powered telemetry system to estimate the postoperative instability of a knee implant.

Estimating in vivo the life span of a total knee replacement prosthesis is currently done by estimating the polyethylene (PE) wear rate from measurement of the femorotibial distance using X-ray photographies. This efficient method requires, however, waiting for few years to obtain a readout. This letter proposes using another metric that can be obtained within a couple of months of surgery, namely the center of pressure (COP). This metric represents the point, where the axial force applies the most onto the tibial tray. The displacement of the COP with respect to its ideal position can be used to estimate the wear and the life span of the PE. This requires the implant to be fitted with a telemetry system described briefly. The proposed method is supported by measures and simulations.

Design and evaluation of instrumented smart knee implant.

The goal of ligament balancing in total knee arthroplasty (TKA) is to distribute the tibiofemoral compressive forces symmetrically between the medial and lateral compartments of a well-aligned prosthetic knee, as well as to reestablish a rectangular and identical tibiofemoral gap in both flexion and extension. Nowadays, the proper alignment of knee mechanical axis and the perfect equalization of flexion and extension gaps are ensured by computer-assisted TKA (CATKA). Nevertheless, any residual imbalance of collateral ligaments at the time of surgery can lead to an excessive imbalance in the postoperative period during the weight-bearing activities, which subject the knee collateral ligaments to increased loading. This in turn leads to an accelerated polyethylene wear, and consequently, to early failure of TKA. The instrumented tibial implant proposed in this study can postoperatively assess and monitor the progression of residual postoperative ligament imbalance of a prosthetic knee, which is perfectly aligned during the surgery thanks to CATKA, via a center-of-pressure (COP)-based approach. This approach depends on the measurement of relative displacement of COP position during the postoperative period with respect to a reference position recorded at the beginning of this period. This measurement is performed for six predetermined flexion angles representative of an entire gait cycle. The tibial implant can also generate the electrical power in addition to their role in monitoring the COP position thanks to the piezoceramics embedded within the tibial tray to achieve this twofold task. Experimental and finite-element analysis (FEA) studies have been conducted to validate the methodology used for the postoperative assessment of residual knee laxity. The issues concerning electrical energy generation and data transmission will be thoroughly discussed in another paper.

Self-powered instrumented knee implant for early detection of postoperative complications.

In-vivo measurement of tibiofemoral forces transmitted through Total Knee Replacement (TKR) during normal walking allows the early detection of postoperative complications such as the tibiofemoral misalignment and soft-tissue imbalance. In addition, the in-vivo data can help to improve the design of TKR in order to reduce polyethylene wear and consequently to increase the lifespan of knee implant. A self-powered custom-designed tibial implant instrumented with four piezoceramics has been developed in order to detect the aforementioned complications by measuring the relative change in pressure center (COP) position for different levels of eccentric compressive loading. Moreover, the energy harvested by the piezoceramics can be used to power a transmission system located at the stem of knee implant to wirelessly transmit the in-vivo data outside the implant for further processing and display.