目的：在放射治療期間，鄰近治療部位之正常組織會接受較低劑量之散射， 為了能準確地評估鄰近組織的吸收劑量，必須事先對熱發光劑量計進行篩選、校正與線性測試。材料與方法：本研究採用 TLD-100H 來進行劑量偵測， 並建立在輻射劑量小於 1 cGy 時，TLD-100H 對低劑量之篩選與校正係數的求法。 同時也求出其劑量反應曲線，以判定計讀電量與照射劑量之間的關係，最後再做不準度測試以驗證其誤差度。結果與討論：由研究的篩選結果可知，本次使用的 25 顆 TLD 中，有 5 顆在預測劑量時，無法將誤差度控制在 3 ％以內，因此將其刪除。由線性度測試可知，劑量在 0 到 0.9 cGy時， 計讀電量與照射劑量之間為線性關係， 且判定係數（ R ?插^為 0.9999。結論：TLD-100H 在低劑量時，有非常高之靈敏度， 並且相當穩定，十分適合低劑量輻射偵測。最後，經過本研究的篩選與校正，TLD-100H 在預測劑量時，可將不準度控制在 3 ％以內。

　　　　　Purpose: It is clinically necessary to evaluate scatter radiation doseto the adjacent normal organs in most radiation therapy procedures. One mustperform the calibration and linearity test for TLD in advance so that thescattered dose can be accurately determined.Material and Method: The TLD-100H is used for dose measurement in the study. Amethod for selection and calibration of TLD in low-level radiation exposure isproposed, and their dose response curves are also determined. To verify thelinear relationship between collected charges and received doses, a blind testis also presented to ensure that the overall uncertainty in under 3% of error.Result: Five out of 25 TLD-100H, failed to report their predict doses within 3%error rate and therefore were eliminated from the study. The test of TLDsverified that the linearity performance is satisfactory with determinationcoefficient of 0.9999 in the range of 0 ～ 0.9 cGy.Conclusion: The TLD-100H is very stable and sensitive, and is especially usefulin low-level radiation detection. If the TLD-100H us carefully selected andcalibrated, then the overall error rate is controlled within 3%, which isapplicable for clinical studies.