We considered whether a positive margin occurring after resection of a soft-tissue sarcoma of a limb would affect the incidence of local recurrence. Patients with low-grade liposarcomas were expected to be a low-risk group as were those who had positive margins planned before surgery to preserve critical structures. Two groups, however, were expected to be at a higher risk, namely, patients who...

We introduce a novel, robust data-driven regularization strategy called Adaptive Regularized Boosting (AR-Boost), motivated by a desire to reduce overfitting. We replace AdaBoost’s hard margin with a regularized soft margin that trades-off between a larger margin, at the expense of misclassification errors. Minimizing this regularized exponential loss results in a boosting algorithm that relaxe...

In the maximum soft margin SVM formulation, penalty is given to each misclassified example, and the influence of the penalty is adjusted by some parameter, say, D. In this paper, based on the geometrical interpretation of this parameter D given in [BB00], we propose a criterion for D and provide an algorithmic way to determine D under this criterion.

We propose a new boosting algorithm for bipartite ranking problems. Our boosting algorithm, called SoftRankBoost, is a modification of RankBoost which maintains only smooth distributions over data. SoftRankBoost provably achieves approximately the maximum soft margin over all pairs of positive and negative examples, which implies high AUC score for future data.

HYPOTHESIS Abdominal wall tumors, though clinically similar, have varying degrees of biological behavior. DESIGN Retrospective review of prospective databases. SETTING Memorial Sloan-Kettering Cancer Center. PATIENTS Eighty-five patients with abdominal wall soft tissue tumors. MAIN OUTCOME MEASURES Primary endpoints included time to first local recurrence, distant metastases, and diseas...

OBJECTIVE The margin of a palatal cleft is a unique anatomical site since the palatal mucosa is continuous with the nasal or nasopharyngeal mucosa. The aim of this study was to compare the expression patterns of cytokeratins and basal membrane components of the mucosa in the area of the cleft. DESIGN Biopsies from the mucosa of the hard palate and from the cleft margin in the soft palate were...

Merkel cell carcinoma is a rare but aggressive neuroendocrine tumor of the skin with high propensity for local, regional and distant soft tissue metastasis not only at initial presentation but also after timely and satisfactory wide-margin surgery. Bone metastases account for 10% of all cases and have been reported involving the calvarium, facial bones and spine related to the head and neck pre...

Kernel-classifiers comprise a powerful class of non-linear decision functions for binary classification. The support vector machine is an example of a learning algorithm for kernel classifiers that singles out the consistent classifier with the largest margin, i.e. minimal real-valued output on the training sample, within the set of consistent hypotheses, the so-called version space. We suggest...

Soft tissue sarcomas have a mortality rate of 40% to 60%, with local recurrence being a poor prognostic factor for overall survival. Three-percent nondiluted hydrogen peroxide is hypothesized to be an effective local adjuvant. The purpose of this study was to identify risk factors for local recurrence in high-grade soft tissue sarcomas and to determine whether using hydrogen peroxide as a local...