| Japanese |
|---|
| Title | 肝シンチグラムの臨床的有効度の定量的評価 - (2) SOL診断の医師間変動の解析 - |
|---|
| Subtitle | 原著 |
|---|
| Authors | 松本徹*, 飯沼武*, 舘野之男*, 町田喜久雄** |
|---|
| Authors(kana) | |
|---|
| Organization | *放射線医学総合研究所臨床研究部, **東京大学医学部放射線科 |
|---|
| Journal | 核医学 |
|---|
| Volume | 19 |
|---|
| Number | 3 |
|---|
| Page | 441-451 |
|---|
| Year/Month | 1982/4 |
|---|
| Article | 原著 |
|---|
| Publisher | 日本核医学会 |
|---|
| Abstract | 「要旨」第1報では肝シンチグラムによるSOL診断の正診率が種々の因子によりどのような影響をうけるかを検討したが, 本報においては, 読影結果の医師間変動を解析することにより, 肝シンチグラム検査の信頼性を定量的に評価した. まず, ROC曲線の医師間変動の程度や傾向を測定し, その結果より医師間変動が小さく, SOL検出効率が最大になる最適動作点がROC曲線上のどの辺に存在するかを示した. 次に各症例に対する読影結果のばらつきから, 「診断の正確さ」や「診断のつけ易さ」が症例の質に依存することを明らかにした. また, 2人の医師の読影のばらつきから読影の個人差の程度を計量し, 最後に, 読影に個人差がある2人の医師をうまく組み合わせると, SOLに対する診断能が向上する可能性があることを示した. |
|---|
| Practice | 臨床医学：一般 |
|---|
| Keywords | interobserver variation, liver image, ROC analysis. |
|---|
| English |
|---|
| Title | Estimation of Clinical Efficacy for Scintigmphic Images of Liver (2) A study on Interobserver Variation of SOL Detection by Image Reading |
|---|
| Subtitle | |
|---|
| Authors | Toru MATUMOTO*, Takeshi A. IINUMA*, Yukio TATENO, Kikuo MACHIDA** |
|---|
| Authors(kana) | |
|---|
| Organization | *National Institute of Radiological Sciences, **Department of Radiology, Tokyo University |
|---|
| Journal | The Japanese Journal of nuclear medicine |
|---|
| Volume | 19 |
|---|
| Number | 3 |
|---|
| Page | 441-451 |
|---|
| Year/Month | 1982/4 |
|---|
| Article | Original article |
|---|
| Publisher | THE JAPANESE SOCIETY OF NUCLEAR MEDICINE |
|---|
| Abstract | The purpose of this study is to investigate the clinical efficacy No. 1(diagnostic accuracy)of liver images on the SOL-disease from a viewpoint of the reliability of diagnosis with liver image. The results of which 11 physicians read the liver images of 401 cases(SOL-presence=124 case, SOL-absence=227 cases)collected from the 8 situations, were analyzed. The reliability of diagnosis in detecting the SOL by reading liver images may be evaluated from the measurements of variation of finding by 11 doctors. The results of analysis are as follows. (1)The amount of interobserver variations of ROC curve in detecting the SOL is different at any operating point. An operating point at which the detection efficiency of SOL-disease is large and the interobserver variations is small, may be searched by calculating the posttest probability differences as a function of the prevalence of disease. (2)By scoring the confidence level of SOL-presence judged by a doctor and calculating an average value and a standard deviation of the scores obtained from 11 doctors, the relation between an average confidence and an interobserver variations of SOL-diagnosis for each case may be estimated. (3)A deviations of answer gained by any two observers out of 11 are calculated by the analysis of variance technique and differences in personality with image reading are detected. (4)Finally, it is presented that the double check study by two doctors having the different personality with image reading improves the ROC curve of SOL-diagnosis. |
|---|
| Practice | Clinical medicine |
|---|
| Keywords | interobserver variation, liver image, ROC analysis. |
|---|