9786139828913 - Development of CADx System for detection of anomalies in breast von Singh, Ankita; Biswas, Mohua; Kale, Rahul (8 Ergebnisse)

Zustand: New.

Taschenbuch. Zustand: Neu. Development of CADx System for detection of anomalies in breast | Ankita Singh (u. a.) | Taschenbuch | 84 S. | Englisch | 2020 | LAP LAMBERT Academic Publishing | EAN 9786139828913 | Verantwortliche Person für die EU: preigu GmbH & Co. KG, Lengericher Landstr. 19, 49078 Osnabrück, mail[at]preigu[dot]de | Anbieter: preigu.

Taschenbuch. Zustand: Neu. This item is printed on demand - it takes 3-4 days longer - Neuware -A dataset of 322 images from Mammographic Image Analysis Society (MIAS). The images are pre-processed with Adaptive Median filter for noise removal and image enhancement. For ROI extraction original image having 1024 x1024 pixels are cropped into 256x256 pixels. The images are segmented using Gaussian Mixture Model for extracting the actual tumour area. We have considered 20 benign, 20 malignant and 20 normal images cases of circumscribed, speculated and ill-defined masses from the database. Each mass is represented with 22 texture features. The PNN classifier is used to perform the classification tasks on 60 images. The input images are classified into three different training -testing datasets as 50-50%,70-30% and 80-20%. With the help of PNN classifier, Sensitivity, Specificity and Accuracy are calculated for each of the dataset. For 50-50% training-testing dataset, Specificity, Sensitivity and Accuracy obtaining are 50%, 50% and 50% respectively. For 70-30% training-testing dataset, Specificity, Sensitivity and Accuracy obtaining are 75%, 100% and 100% respectively. 80-20% training-testing dataset gave the promising results with the Specificity, Sensitivity and Accuracy of 100% 84 pp. Englisch.

Zustand: New. Print on Demand.

Zustand: New. PRINT ON DEMAND.

Zustand: New. Dieser Artikel ist ein Print on Demand Artikel und wird nach Ihrer Bestellung fuer Sie gedruckt. Autor/Autorin: Singh AnkitaAnkita Singh Mohua BiswasRahul KaleA dataset of 322 images from Mammographic Image Analysis Society (MIAS). The images are pre-processed with Adaptive Median filter for noise removal and image enhancement. For ROI ex.

Taschenbuch. Zustand: Neu. This item is printed on demand - Print on Demand Titel. Neuware -A dataset of 322 images from Mammographic Image Analysis Society (MIAS). The images are pre-processed with Adaptive Median filter for noise removal and image enhancement. For ROI extraction original image having 1024 x1024 pixels are cropped into 256x256 pixels. The images are segmented using Gaussian Mixture Model for extracting the actual tumour area. We have considered 20 benign, 20 malignant and 20 normal images cases of circumscribed, speculated and ill-defined masses from the database. Each mass is represented with 22 texture features. The PNN classifier is used to perform the classification tasks on 60 images. The input images are classified into three different training -testing datasets as 50-50%,70-30% and 80-20%. With the help of PNN classifier, Sensitivity, Specificity and Accuracy are calculated for each of the dataset. For 50-50% training-testing dataset, Specificity, Sensitivity and Accuracy obtaining are 50%, 50% and 50% respectively. For 70-30% training-testing dataset, Specificity, Sensitivity and Accuracy obtaining are 75%, 100% and 100% respectively. 80-20% training-testing dataset gave the promising results with the Specificity, Sensitivity and Accuracy of 100%VDM Verlag, Dudweiler Landstraße 99, 66123 Saarbrücken 84 pp. Englisch.

Taschenbuch. Zustand: Neu. nach der Bestellung gedruckt Neuware - Printed after ordering - A dataset of 322 images from Mammographic Image Analysis Society (MIAS). The images are pre-processed with Adaptive Median filter for noise removal and image enhancement. For ROI extraction original image having 1024 x1024 pixels are cropped into 256x256 pixels. The images are segmented using Gaussian Mixture Model for extracting the actual tumour area. We have considered 20 benign, 20 malignant and 20 normal images cases of circumscribed, speculated and ill-defined masses from the database. Each mass is represented with 22 texture features. The PNN classifier is used to perform the classification tasks on 60 images. The input images are classified into three different training -testing datasets as 50-50%,70-30% and 80-20%. With the help of PNN classifier, Sensitivity, Specificity and Accuracy are calculated for each of the dataset. For 50-50% training-testing dataset, Specificity, Sensitivity and Accuracy obtaining are 50%, 50% and 50% respectively. For 70-30% training-testing dataset, Specificity, Sensitivity and Accuracy obtaining are 75%, 100% and 100% respectively. 80-20% training-testing dataset gave the promising results with the Specificity, Sensitivity and Accuracy of 100%.