| Japanese |
|---|
| Title | 容量可変心室ファントムおよび臨床症例における左心室輪郭抽出法の検討 |
|---|
| Subtitle | 原著 |
|---|
| Authors | 中嶋憲一*, 山田正人*, 分校久志*, 多田明*, 利波紀久*, 久田欣一* |
|---|
| Authors(kana) | |
|---|
| Organization | *金沢大学医学部核医学教室, **金沢大学医学部附属病院アイソトープ部 |
|---|
| Journal | 核医学 |
|---|
| Volume | 18 |
|---|
| Number | 10 |
|---|
| Page | 1405-1411 |
|---|
| Year/Month | 1981/12 |
|---|
| Article | 原著 |
|---|
| Publisher | 日本核医学会 |
|---|
| Abstract | 「要旨」容積を自由に変化できる心室ファントムを作製し, 拡張末期カウント(EDC), 収縮末期カウント, (ESC), 駆出分画に(EF)ついて3種類の輪郭抽出法すなわち微分法, threshold法, 固定ROI法を比較検討した. EFは微分法では理論値より高値に, 後2者では低値に算出された. EDCとESCは微分法では再現性が良いが, threshold法を用いる場合はバッグラウンド減算後, 最大カウントの30%の辺縁が適切であった. 臨床例23例で微分法とthreshold法によるEFを左心室造影結果と比較すると, 相関は両者で差がなかった(r=0.82)が, 一次回帰を行うと微分法は相対的に高値, threshold法は低値であった. データ収集時間を30秒, 1, 2, 5, 15分間としEDC, EFを検討すると, 微分法, threshold法ともに1分間のデータ収集でも満足できる変動であり, 運動負荷など低計数の場合にも有効と考えられた. |
|---|
| Practice | 臨床医学：一般 |
|---|
| Keywords | Edge detection, Ventricular phantom, Gated blood pool study, Data acquisition time |
|---|
| English |
|---|
| Title | Validity of Left Ventricular Edge - Detection Methods in Ventricular Phantom and Clinical Cases |
|---|
| Subtitle | Original article |
|---|
| Authors | Ken'ichi NAKAJIMA*, Masato YAMADA**, Hisashi BUNKO*, Akira TADA*, Norihisa TONAMI*, Kin'ichi HISADA* |
|---|
| Authors(kana) | |
|---|
| Organization | *Department of Nuclear Medicine, School of Medicine, Kanazawa University, **Division of Central Isotope Service, Kanazawa University Hospital |
|---|
| Journal | The Japanese Journal of nuclear medicine |
|---|
| Volume | 18 |
|---|
| Number | 10 |
|---|
| Page | 1405-1411 |
|---|
| Year/Month | 1981/12 |
|---|
| Article | Original article |
|---|
| Publisher | THE JAPANESE SOCIETY OF NUCLEAR MEDICINE |
|---|
| Abstract | Three ventricular edge-detection methods, i. e. differentiation, thresholding and fixed region-of-interest(ROI)techniques were evaluated as to end-diastolic count (EDC), end-systolic count(ESC)and ejection fraction(EF)using a ventricular phantom. In the differentiation method, the reproducibilities of EDC and ESC were the best, but EF was estimated higher than true EF. In the thresholding method, 30% threshold of the maximum counts after background(BKG)subtraction was supposed to be appropriate. However, EFs were calcualted lower than true EF. In the fixed ROI method, EF was estimated the lowest. In the differentiation technique EF and ESC were least affected by the change of BKG. Correlations between the nuclear EFs and contrast EFs calculated from the left ventriculography were compared in 23 cases. EFs obtained by the differentiation method were higher(r=0.82) and those obtained by the thresholding method were lower(r=0.82)than the contrast method. There was no difference of correlation coefficients between the contrast EF and each nuclear EF. Using short acquisition times of gated blood pool data, EF and EDC were calculated. Even when 1 min. acquisition time was used, EF and EDC were reliable both by differentiation and thresholding techniques. |
|---|
| Practice | Clinical medicine |
|---|
| Keywords | Edge detection, Ventricular phantom, Gated blood pool study, Data acquisition time |
|---|