以模糊推論方法為基礎之胎兒不保證狀態監測系統

胎兒監測在產檢過程中一向是醫師判斷的重要依據，根據監測系統紀錄的胎兒心跳速率(FHR)與孕婦子宮壓力(UP)數據，可提前發現許多可能造成胎兒先天疾病與無法順產的徵兆，並給予適當治療與醫療處置。然傳統的儀器設備雖提供準確的監測數值，卻缺少個人化與自動分析的功能，因而專業醫護人員需以人力判讀長時間累積的數據，也可能產生因人而異的判斷標準；本論文即以自動化分析胎兒心跳速率與孕婦子宮壓力數據為目標，建置一套用模糊推論方法分析的不保證胎兒狀態監測系統，醫師根據受測者的身體狀況設定個人參數值(例如：體溫)，接收到訊號後，透過加權平均的統計方法估算兩者訊號之基線值，再利用模糊推論方法加以分析訊號，找出監測時段中已定義的不保證胎兒狀態(non-reassuring fetal status)。過程中，醫護人員可在系統警告鈴聲響起時給予適當的醫療處置，而儲存載入功能也提供醫療上的追蹤與之後監測延續為用。鑒於目前醫療用儀器之限制，我們建立一個胎兒心跳速率與孕婦子宮壓力的訊號模擬器，以測試監測系統的可行性與可信度。透過專業臨床醫師的鑑定，我們的監測系統可上達九成五的準確率。

關鍵字

不保證胎兒狀態；模糊推論方法；胎兒心跳速率；孕婦子宮壓力

並列摘要

Fetal heart rate (FHR) and uterine pressure (UP) are two of the most important statistics in antenatal examination. Safe and steady fetal monitoring signals lead to non-risky fetal status. Traditionally to achieve such a status, obstetricians check fetal heart rate (FHR) and uterine pressure (UP) signals manually, and diagnose the probable status of fetus. This manual processing of such ultrasonic data takes time and labor. To overcome this problem, we proposed a cheaper and more efficient fetal non-reassuring status monitoring system to help obstetricians detect non-reassuring fetal status. At first the weighted average is employed to estimate the fetal heartbeat baseline and uterine contraction baseline, then the system recognizes heartbeat acceleration, heartbeat deceleration, uterine construction, heartbeat noise pattern, and uterine noise pattern based on those baselines. Moreover, the monitoring system considers five types of non-reassuring fetal status. Fuzzy logic is used to analyze the signals for each non-reassuring status type and 26 fuzzy rules are used to recognize non-reassuring fetal status that triggers an alarm mechanism. When non-reassuring status is found out, the alarm mechanism will be triggered for immediate treatment. We make this monitoring system modifiable and adoptable to fit the requirements of specific patients. For verification, a signal simulator is designed. The experimental results show the accurate rate can reach 95%, when our system is examined by the obstetricians.

並列關鍵字

Non-reassuring fetal status ； fuzzy inference method ； fetal heart rate ； uterine pressure

參考文獻

[2]O. Fontenla-Romero, A. Alonso-Betanzos and B. Guijarro-Berdinas, “Adaptive pattern recognition in the analysis of cardiotocographic records,” IEEE Tran. on Neural Networks, pp.1188-1195, Sept. 2001.

[3]Y.-P. Huang and C.-H. Yang, “Using fuzzy inference method to design a fetal distress monitoring system,” Proc. of Ninth Conf. on Artificial Intelligence and Applications, Taipei, Taiwan, pp.1-6, Nov. 2004.

[4]M.I. Ibrahimy, F. Ahmed, M.A.M Ali and E. Zahedi, “Real-time signal processing for fetal heart rate monitoring,” IEEE Tran. on Biomedical Engineering, pp.258-261, Feb. 2003.

[5]L. Jimenez, R. Gonzalez, M. Gaitan, S. Carrasco and C. Vargas, “Computerized algorithm for baseline estimation of fetal heart rate,” Computers in Cardiology, pp.477-480, Sept. 2002.

[6]F. Kovacs, M. Torok and I. Habermajer, “A rule-based phonocardiographic method for long-term fetal heart rate monitoring,” IEEE Trans. on Biomedical Engineering, pp.124-130, Jan. 2000.

國際替代計量

以模糊推論方法為基礎之胎兒不保證狀態監測系統

未授權

主題瀏覽