Dataset Information

A Directed Acyclic Graph-Large Margin Distribution Machine Model for Music Symbol Classification.

ABSTRACT: Optical Music Recognition (OMR) has received increasing attention in recent years. In this paper, we propose a classifier based on a new method named Directed Acyclic Graph-Large margin Distribution Machine (DAG-LDM). The DAG-LDM is an improvement of the Large margin Distribution Machine (LDM), which is a binary classifier that optimizes the margin distribution by maximizing the margin mean and minimizing the margin variance simultaneously. We modify the LDM to the DAG-LDM to solve the multi-class music symbol classification problem. Tests are conducted on more than 10000 music symbol images, obtained from handwritten and printed images of music scores. The proposed method provides superior classification capability and achieves much higher classification accuracy than the state-of-the-art algorithms such as Support Vector Machines (SVMs) and Neural Networks (NNs).

SUBMITTER: Wen C

PROVIDER: S-EPMC4795649 | biostudies-literature | 2016

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

A Directed Acyclic Graph-Large Margin Distribution Machine Model for Music Symbol Classification.

Wen Cuihong C Zhang Jing J Rebelo Ana A Cheng Fanyong F

PloS one 20160317 3

Optical Music Recognition (OMR) has received increasing attention in recent years. In this paper, we propose a classifier based on a new method named Directed Acyclic Graph-Large margin Distribution Machine (DAG-LDM). The DAG-LDM is an improvement of the Large margin Distribution Machine (LDM), which is a binary classifier that optimizes the margin distribution by maximizing the margin mean and minimizing the margin variance simultaneously. We modify the LDM to the DAG-LDM to solve the multi-cla ...[more]

PMID: 26985826

Similar Datasets

Project description:BackgroundHigh tibial osteotomy (HTO) is increasingly used in young and physically active patients with knee osteoarthritis. These patients have high expectations, including return to sport (RTS). By retaining native knee structures, a return to highly knee-demanding activities seems possible. However, evidence on patient-related outcomes, including RTS, is sparse. Also, time to RTS has never been described. Furthermore, prognostic factors for RTS after HTO have never been investigated. These data may further justify HTO as a surgical alternative to knee arthroplasty.PurposeTo investigate the extent and timing of RTS after HTO in the largest cohort investigated for RTS to date and to identify prognostic factors for successful RTS.Study designCase-control study; Level of evidence, 3.MethodsConsecutive patients with HTO, operated on between 2012 and 2015, received a questionnaire. First, pre- and postoperative sports participation questions were asked. Also, time to RTS, sports level and frequency, impact level, the presymptomatic and postoperative Tegner activity score (1-10; higher is more active), and the postoperative Lysholm score (0-100; higher is better) were collected. Finally, prognostic factors for RTS were analyzed using a logistic regression model. Covariates were selected based on univariate analysis and a directed acyclic graph.ResultsWe included 340 eligible patients of whom 294 sufficiently completed the questionnaire. The mean follow-up was 3.7 years (± 1.0 years). Out of 256 patients participating in sports preoperatively, 210 patients (82%) returned to sport postoperatively, of whom 158 (75%) returned within 6 months. We observed a shift to participation in lower-impact activities, although 44% of reported sports activities at final follow-up were intermediate- or high-impact sports. The median Tegner score decreased from 5.0 (interquartile range [IQR], 4.0-6.0) presymptomatically to 4.0 (IQR, 3.0-4.0) at follow-up (P < .001). The mean Lysholm score at follow-up was 68 (SD, ± 22). No significant differences were found between patients with varus or valgus osteoarthritis. The strongest prognostic factor for RTS was continued sports participation in the year before surgery (odds ratio, 2.81; 95% CI, 1.37-5.76).ConclusionMore than 8 of 10 patients returned to sport after HTO. Continued preoperative sports participation was associated with a successful RTS. Future studies need to identify additional prognostic factors.

Dataset Information

A Directed Acyclic Graph-Large Margin Distribution Machine Model for Music Symbol Classification.

Publications

A Directed Acyclic Graph-Large Margin Distribution Machine Model for Music Symbol Classification.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets