光聲多層膜及其於全學式超音波/光聲影像探頭之應用

Abstract

超音波與光聲多模態影像在生物醫學影像上已有相當多的研究，其能有效率的將兩種不同性質的影像結合。然而一如許多其他種類的多模態影像，超音波/光聲多模態影像因其個別影像的能量來源不同，超音波與光聲信號必須分開產生，造成影像對位以及時間上的排列等問題，進而限制了某些特定的應用。在本研究中我們提出一種能在單發雷射內同時產生超音波以及光聲影像的方法，此方法利用吸收通過光聲多層膜的雷射光來產生超音波形成超音波影像，並使用同一發雷射所剩餘的能量照射組織產生光聲影像，如此一來即能在單發雷射內同時產生超音波及光聲影像，解決傳統超音波與光聲多模態影像之影像對位以及時間上的排列問題。在之前的研究中也有利用過類似的架構，不過之前的研究是利用單層膜來產生光產生超音波信號，而此信號具有相當寬頻的特性，使得在頻域無法上分離接收回來的較低頻光聲以及光產生之超音波信號，因此在擷取超音波/光聲多模態影像時，超音波與光聲信號依然必須分開產生。本研究所使用的光學多層膜能產生較窄頻的超音波信號，因此在頻譜上能與低頻的光聲信號進行分離。此光學吸收體中有多層的光吸收層與光穿透層，當改變光吸收層之間的距離以及吸收係數時，即可調整信號產生之頻率特性。本研究在實作上成功的利用光聲多層膜產生了中心頻率14~28 MHz不等的超音波訊號，並且相較於單層膜產生之相對頻寬約100%的超音波信號，依此方法產生的高頻窄頻光學超音波信號相對頻寬約為30%，能避開生物組織之光聲信號約10MHz的低頻頻段，因此在影像處理上能利用部分頻帶法以提高超音波與光聲影像之對比。除此之外本研究在架構上進一步的搭配聚合物微環之光學式超音波接收器，形成全光學式超音波/光聲探頭架構。在實驗架構上，我們使用此超音波/光聲探頭架構截取囊腫狀仿體以及薄膜仿體之超音波回波以及光聲信號以測試此架構之可行性，並且利用合成孔徑演算法來提高影像的橫向解析度以及訊雜比。

Multimodality ultrasound (US) and photoacoustic (PA) imaging has received wide research attention for biomedical research. It is an efficient way to combine two complementary imaging mechanisms within one single system. Nevertheless, like many other multimodality imaging systems, generally different energy sources and detectors need to be used. Thus, image registration and temporal alignment become a critical issue limiting the imaging performance in certain applications. In this study, we propose a new imaging method that requires only a single laser pulse to concurrently perform US and PA imaging. Specifically, we propose a thin film with multiple optically absorbing layers which is also partially transparent to light. With this multilayer thin film, the transmission light can be used to perform PA imaging, whereas the absorbed light energy can generate US for US imaging. As the PA and US images are created by the same laser pulse, the image registration and temporal alignment problems practically no longer exist. In our previous study, a similar approach was taken but with a single layer film. Because of the broadband nature of the generated US, the US signal and the PA signal are spectrally overlapped, thus making it difficult to be separated. By using the multilayer film proposed in the current study, the generated US signal has a relatively narrower bandwidth and thus it is spectrally separable from the PA signal. Characteristics of the generated US signal can also be tuned by adjusting the optical absorption coefficient of light-absorbing layers as well as thickness of the layers. In our designs, the US signals generated by the multilayer films typically have the center frequency ranging from 14 to 28MHz. The bandwidth is typically around 30%, compared to the 100% bandwidth from the single layer films. The PA signal, on the other hand, generally is most sensitive around 10MHz or lower for biological tissues. Thus, the US signal and the PA signal can be separated using a filter at the receiver end. In addition, when using a microring to detect both the US signal and the PA signal, the US and PA imaging system becomes all optical. A thin film phantom and a cyst-like phantom were used to test imaging performance of this approach. The feasibility is demonstrated. The lateral resolution and SNR can be further improved by applying the synthetic focusing technique.