Abstract
This paper presents the results of a CMOS-VLSI implementation of a realistic computational model proposed by Walter Freeman for the olfactory system. This model, in later years, has been studied for engineering applications such as auto-association and classification. The analogue nature of the model motivates analogue VLSI implementations. However, the dimension and complexity of such system poses many obstacles to an analogue electronic implementation; one such is the massive interconnectivity which size increases with the square of the number of inputs (channels). We suggest a multiplexing procedure that puts the burden of interconnectivity over a digital system that is simpler to design and makes the analogue system more treatable. The procedure naturally samples the signals. To avoid smoothing filters, a discrete-time solution was also employed. Although with such approach the time resolution is reduced, the advantages overcome the detriments. Previous work has shown that the model can be efficiently discretized using DSP techniques, resulting on a system that is able to predict, on sample-by-sample basis, the behaviour of the VLSI circuit, allowing for a simple and flexible way to adjust the circuit parameters. We present the measured circuit results that are further confronted with the digital implementation.

Abstract
Byte stuffing is a technique to allow the transparent transmission of arbitrary sequences with constrained sequences. To date, most of the existing algorithms, such as PPP, attain a low average overhead by sacrificing the worst-case scenario. An exception is COBS which was designed for a low worst-case overhead; however, it imposes always a nonzero overhead, even on small packets. In this work is proposed a byte stuffing algorithm that simultaneously controls the average and worst-case overhead, performing close to the theoretical bound. It is shown analytically that the proposed algorithm achieves improved average and worst-case rates over state of the art methods. Furthermore, this technique is generalized to hybrid methods, with lower computing complexity. It is further analysed and compared experimentally the behaviour of the proposed algorithm against established algorithms in terms of byte overhead and computational time.

Abstract
Classification of ordinal data is one of the most important tasks of relation learning. This paper introduces a new machine learning paradigm specifically intended for classification problems where the classes have a natural order. The technique reduces the problem of classifying ordered classes to the standard two-class problem. The introduced method is then mapped into support vector machines and neural networks. Generalization bounds of the proposed ordinal classifier are also provided. An experimental study with artificial and real data sets, including an application to gene expression analysis, verifies the usefulness of the proposed approach.

Abstract
Objective: This work presents a novel approach for the automated prediction of the aesthetic result of breast cancer conservative treatment (BCCT). Cosmetic assessment plays a major rote in the study of BCCT. Objective assessment methods are being preferred to overcome the drawbacks of subjective evaluation. Methodology: The problem is addressed as a pattern recognition task. A dataset of images of patients was classified in four classes (excellent, good, fair, poor) by a panel of international experts, providing a gold standard classification. As possible types of objective features we considered those already identified by domain experts as relevant to the aesthetic evaluation of the surgical procedure, namely those assessing breast asymmetry, skin colour difference and scar visibility. A classifier based on support vector machines was developed from objective features extracted from the reference dataset. Results: A correct classification rate of about 70% was obtained when categorizing a set of unseen images into the aforementioned four classes. This accuracy is comparable with the result of the best evaluator from the panel of experts. Conclusion: The results obtained are rather encouraging and the developed tool, could be very helpful in assuring objective assessment of the aesthetic outcome of BCCT.

Abstract

Abstract
The aim of this study was to evaluate the factors that determine esthetic outcome after breast cancer conservative treatment, based on a consensual classification obtained with an international consensus panel. Photographs were taken from 120 women submitted to conservative unilateral breast cancer surgery (with or without axillary surgery) and radiotherapy. The images were sent to a panel of observers from 13 different countries and consensus on the classification of esthetic result (recorded as excellent, good, fair or poor) was obtained in 113 cases by means of a Delphi method. For each patient, data were collected retrospectively regarding patient characteristics, tumor, and treatment factors. Univariate and multivariate analysis were used to evaluate the correlation between these factors and overall cosmetic results. On univariate analysis, younger and thinner patients as well as patients with lower body mass index (BMI) and premenopausal status obtained better cosmetic results. In the group of tumor- and treatment-related factors, larger removed specimens, clearly visible scars, the use of chemotherapy and longer follow-up period were associated with less satisfactory results. On multivariate analysis, only BMI and scar visibility maintained a significant association with cosmesis. BMI and scar visibility are the only factors significantly associated with cosmetic results of breast cancer conservative treatment, as evaluated by an international consensus panel.