| Presentation | 1994/1/27 Diagnosis support in mental medicine using probabilistic Logic neural network:mSDN versus AI and Bayesian statis tics. Kazuhiro Iida, Kazuo Yana, Kohji Kawachi, Yoshiaki Ohkubo, Fumio Okuyama, Michio Tohru, |
|---|---|
| PDF Download Page | PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | To refine question items and to support diagnosis,100 of psychological questionnaires are analyzed using an AI algorithm ID3,Bayesian statistics,and the mSDN,which can extract comprehensible rules from the data.Bayesian statistics and the rules extracted by the mSDN predict final diagnosis with the highest accuracies,which are 10 to 20% higher than those of a real doctor.The rules by the mSDN are most robust against changing question items.The mSDN rules,which are two times simpler than ID3 rules,are judged reasonable by a psychiatrist.These results suggest that the mSDN can be a useful tool in making good questionnaires,and that this network has advantages over other techniques. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Neural Networks / Diagnosis / probabilistic Logic / Rule- extraction / Bayesian statistics / ID3 |
| Paper # | MBE93-108 |
| Date of Issue |
| Conference Information | |
| Committee | MBE |
|---|---|
| Conference Date | 1994/1/27(1days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | |
| Vice Chair | |
| Secretary | |
| Assistant | |
| Paper Information | |
| Registration To | ME and Bio Cybernetics (MBE) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Diagnosis support in mental medicine using probabilistic Logic neural network:mSDN versus AI and Bayesian statis tics. |
| Sub Title (in English) | |
| Keyword(1) | Neural Networks |
| Keyword(2) | Diagnosis |
| Keyword(3) | probabilistic Logic |
| Keyword(4) | Rule- extraction |
| Keyword(5) | Bayesian statistics |
| Keyword(6) | ID3 |
| 1st Author's Name | Kazuhiro Iida |
| 1st Author's Affiliation | Fundamental Research Laboratories,NEC Corporation() |
| 2nd Author's Name | Kazuo Yana |
| 2nd Author's Affiliation | Department of Electrical Engineering,Hosei University |
| 3rd Author's Name | Kohji Kawachi |
| 3rd Author's Affiliation | Department of Electrical Engineering,Hosei University |
| 4th Author's Name | Yoshiaki Ohkubo |
| 4th Author's Affiliation | Faculty of Medicine,Tokyo Medical Dental University |
| 5th Author's Name | Fumio Okuyama |
| 5th Author's Affiliation | Faculty of Medicine,Tokyo Medical Dental University Hospital |
| 6th Author's Name | Michio Tohru |
| 6th Author's Affiliation | Faculty of Medicine,Tokyo Medical Dental University |
| Date | 1994/1/27 |
| Paper # | MBE93-108 |
| Volume (vol) | vol.93 |
| Number (no) | 448 |
| Page | pp.pp.- |
| #Pages | 8 |
| Date of Issue | |