Name
Abstract | ||
|---|---|---|
The early diagnosis of a cancer type is a fundamental goal in cancer treatment, as it can facilitate the subsequent clinical management of patients. The leading importance of classifying cancer patients into high or low risk groups has led many research teams, both from biomedical and bioinformatics field, to study the application of Deep Learning (DL) methods. The ability of DL tools to detect key features from complex datasets is a fundamental achievement in early diagnosis and cell cancer progression. In this paper, we apply DL approach to classification of osteosarcoma cells. Osteosarcoma is the most common bone cancer occurring prevalently in children or young adults. Glass slides of different cell populations were cultured from Mesenchimal Stromal Cells (MSCs) and differentiated in healthy bone cells (osteoblasts) or osteosarcoma cells. Images of such samples are recorded with an optical microscope. DL is then applied to identify and classify single cells. The results show a classification accuracy of 0.97. The next step is the application of our DL approach to tissue in order to improve digital histopathology. |
Year | DOI | Venue |
|---|---|---|
2018 | 10.1007/978-3-319-98678-4_36 | MULTIMEDIA AND NETWORK INFORMATION SYSTEMS |
Keywords | DocType | Volume |
Osteosarcoma cells,Deep Learning,Convolutional neural networks,Convolutional object detection systems,Cell classification | Conference | 833 |
ISSN | Citations | PageRank |
2194-5357 | 0 | 0.34 |
References | Authors | |
0 | 3 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Mario D'Acunto | 1 | 10 | 4.03 |
| Massimo Martinelli | 2 | 47 | 10.09 |
| Davide Moroni | 3 | 63 | 20.21 |