AIM To evaluate the efficacy of doubling time(DT) of gastrointestinal submucosal tumors(GIST).METHODS From April 1987 through November 2012, a total of 323 patients were given a final histopathological diagnosis of GI...AIM To evaluate the efficacy of doubling time(DT) of gastrointestinal submucosal tumors(GIST).METHODS From April 1987 through November 2012, a total of 323 patients were given a final histopathological diagnosis of GISTs on surgical resection or endoscopic ultrasound-guided fine-needle aspiration(EUS-FNA) in Kitasato University East Hospital or Kitasato University Hospital. We studied 53 of these patients(34 with resected tumors and 19 with unresected tumors) whose tumors could be measured on EUS on at least two successive occasions. The histopathological diagnosis was GIST in 34 patients, leiomyoma in 5, schwannoma in 3, ectopic pancreas in 1, hamartoma in 1, cyst in 1, Brunner's adenoma in 1, and spindle-cell tumor in 7. We retrospectively calculated the DT of GISTs on the basis of the time course of EUS findings to estimate the growth rate of such tumors.RESULTS The DT was 17.2 mo for GIST, as compared with 231.2 mo for leiomyoma, 104.7 mo for schwannoma, 274.9mo for ectopic pancreas, 61.2 mo for hamartoma, 49.0 mo for cyst, and 134.7 mo for Brunner's adenoma. The GISTs were divided into risk classes on the basis of tumor diameters and mitotic figures(Fletcher's classification). The classification was extremely low risk or low risk in 28 patients, intermediate risk in 3, and high risk in 3. DT of GIST according to risk was 24.0 mo for extremely low-risk plus low-risk GIST, 17.1 mo for intermediate-risk GIST, and 3.9 mo for high-risk GIST. DT of GIST was significantly shorter than that of leiomyoma plus schwannoma(P < 0.05), and DT of high-risk GIST was significantly shorter than that of extremely low-risk plus low-risk GIST(P < 0.05).CONCLUSION For GIST, a higher risk grade was associated with a significantly shorter DT. Small SMTs should initially be followed up within 6 mo after detection.展开更多
In this paper,we propose a new algorithm to extend support vector machine(SVM)for binary classification to multicategory classification.The proposed method is based on a sequential binary classification algorithm.We f...In this paper,we propose a new algorithm to extend support vector machine(SVM)for binary classification to multicategory classification.The proposed method is based on a sequential binary classification algorithm.We first classify a target class by excluding the possibility of labeling as any other classes using a forward step of sequential SVM;we then exclude the already classified classes and repeat the same procedure for the remaining classes in a backward step.The proposed algorithm relies on SVM for each binary classification and utilizes only feasible data in each step;therefore,the method guarantees convergence and entails light computational burden.We prove Fisher consistency of the proposed forward–backward SVM(FB-SVM)and obtain a stochastic bound for the predicted misclassification rate.We conduct extensive simulations and analyze real-world data to demonstrate the superior performance of FB-SVM,for example,FB-SVM achieves a classification accuracy much higher than the current standard for predicting conversion from mild cognitive impairment to Alzheimer’s disease.展开更多
文摘AIM To evaluate the efficacy of doubling time(DT) of gastrointestinal submucosal tumors(GIST).METHODS From April 1987 through November 2012, a total of 323 patients were given a final histopathological diagnosis of GISTs on surgical resection or endoscopic ultrasound-guided fine-needle aspiration(EUS-FNA) in Kitasato University East Hospital or Kitasato University Hospital. We studied 53 of these patients(34 with resected tumors and 19 with unresected tumors) whose tumors could be measured on EUS on at least two successive occasions. The histopathological diagnosis was GIST in 34 patients, leiomyoma in 5, schwannoma in 3, ectopic pancreas in 1, hamartoma in 1, cyst in 1, Brunner's adenoma in 1, and spindle-cell tumor in 7. We retrospectively calculated the DT of GISTs on the basis of the time course of EUS findings to estimate the growth rate of such tumors.RESULTS The DT was 17.2 mo for GIST, as compared with 231.2 mo for leiomyoma, 104.7 mo for schwannoma, 274.9mo for ectopic pancreas, 61.2 mo for hamartoma, 49.0 mo for cyst, and 134.7 mo for Brunner's adenoma. The GISTs were divided into risk classes on the basis of tumor diameters and mitotic figures(Fletcher's classification). The classification was extremely low risk or low risk in 28 patients, intermediate risk in 3, and high risk in 3. DT of GIST according to risk was 24.0 mo for extremely low-risk plus low-risk GIST, 17.1 mo for intermediate-risk GIST, and 3.9 mo for high-risk GIST. DT of GIST was significantly shorter than that of leiomyoma plus schwannoma(P < 0.05), and DT of high-risk GIST was significantly shorter than that of extremely low-risk plus low-risk GIST(P < 0.05).CONCLUSION For GIST, a higher risk grade was associated with a significantly shorter DT. Small SMTs should initially be followed up within 6 mo after detection.
基金This work is supported by NIH Grants R01GM124104,NS073671,NS082062,NUL1 RR025747Alzheimer’s Disease Neuroimaging Initiative(ADNI)(U01 AG024904,DOD ADNI,W81XWH-12-2-0012),and a pilot award from the Gillings Innovation Lab at the University of North Carolina.The authors acknowledge the investigators within the ADNI who contributed to the design and implementation of ADNI.
文摘In this paper,we propose a new algorithm to extend support vector machine(SVM)for binary classification to multicategory classification.The proposed method is based on a sequential binary classification algorithm.We first classify a target class by excluding the possibility of labeling as any other classes using a forward step of sequential SVM;we then exclude the already classified classes and repeat the same procedure for the remaining classes in a backward step.The proposed algorithm relies on SVM for each binary classification and utilizes only feasible data in each step;therefore,the method guarantees convergence and entails light computational burden.We prove Fisher consistency of the proposed forward–backward SVM(FB-SVM)and obtain a stochastic bound for the predicted misclassification rate.We conduct extensive simulations and analyze real-world data to demonstrate the superior performance of FB-SVM,for example,FB-SVM achieves a classification accuracy much higher than the current standard for predicting conversion from mild cognitive impairment to Alzheimer’s disease.
