I-Corps: Artificial Intelligence (AI)-based Image Fusion Technology for Guiding Prostate Biopsies

然而，只有不到20%的手术是在融合指导下进行的。硬件设置起来很麻烦，并且大大增加了系统的成本。现有的系统还要求临床用户手动对准MRI和超声波的体积，这称为图像配准。图像配准是一项具有挑战性的任务，对系统的精度有重大影响，往往导致此类系统的性能较差。提出的技术使用人工智能模型来消除对跟踪硬件的需求，并降低图像配准的专业要求。这项技术可以通过提高患者吞吐量，最大限度地减少系统设置时间，并增加即使在设备较少的临床中也可以进行融合引导活检的可及性，从而改进活检程序。该i-Corps项目基于使用人工智能(AI)的图像融合技术的开发，旨在消除癌症活检程序中使用的外部跟踪硬件的需求。所提出的技术利用人工智能和机器学习来自动化融合过程，从而产生更高效和更准确的活检引导系统。它使用人工智能算法自动识别输入图像中的图像特征，用于体积重建、跨通道图像配准和帧到体积映射。这项技术工作已经被记录在同行评议的论文、专利申请和回顾研究中，以验证这项技术。此外，拟议的技术解决了医生面临的挑战，提供了提高活检准确性、降低成本和缩短学习曲线的机会。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

The broader impact/commercial potential of this I-Corps project is the development of an artificial intelligence (AI)-driven image fusion technology designed to guide cancer biopsies. Currently, the gold standard approach for guiding biopsies employs fusing magnetic resonance imaging (MRI) and ultrasound, which can increase cancer biopsy yield by 30%. However, less than 20% of procedures are performed with fusion guidance. The existing methods for fusing MRI and ultrasound images require external tracking hardware that acts like a “medical GPS” to track the position and orientation of an ultrasound probe. The hardware is cumbersome to set up and adds significant cost to the system. The available systems also require clinical users to manually align MRI and ultrasound volumes, which is called image registration. Image registration is a challenging task and has a significant impact on the accuracy of the system, often leading to the poor performance of such systems. The proposed technology uses AI models to remove the need for tracking hardware and reduce the expertise requirement for image registration. This technology may improve biopsy procedures by enhancing patient throughput, minimizing system setup time, and increasing the accessibility of fusion-guided biopsies even in less-equipped clinics.This I-Corps project is based on the development of image fusion technology using Artificial Intelligence (AI) that aims to eliminate the need for external tracking hardware used in cancer biopsy procedures. The proposed technology utilizes AI and machine learning to automate the fusion process, resulting in a more efficient and accurate biopsy guidance system. It uses AI algorithms to automatically identify image features within the input images for volume reconstruction, cross modality image registration, and frame to volume mapping. Results comparing his technology with traditional methods that depend heavily on external tracking devices for fusing magnetic resonance imaging (MRI) and ultrasound images show that it reduces human intervention and improves image fusion performance. The technical work has been documented in peer-reviewed papers, patent filings, and retrospective studies to validate the technology. In addition, the proposed technology addresses the challenges faced by doctors, offering an opportunity for improved biopsy accuracy, reduced costs, and a shortened learning curve.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.