Enhancing Ultrasound Image Quality Across Disease Domains: Application of Cycle-Consistent Generative Adversarial Network and Perceptual Loss

Abstract

Background: Numerous studies have explored image processing techniques aimed at enhancing ultrasound images to narrow the performance gap between low-quality portable devices and high-end ultrasound equipment. These investigations often use registered image pairs created by modifying the same image through methods like down sampling or adding noise, rather than using separate images from different machines. Additionally, they rely on organ-specific features, limiting the models' generalizability across various imaging conditions and devices. The challenge remains to develop a universal framework capable of improving image quality across different devices and conditions, independent of registration or specific organ characteristics.

Objective: This study aims to develop a robust framework that enhances the quality of ultrasound images, particularly those captured with compact, portable devices, which are often constrained by low quality due to hardware limitations. The framework is designed to effectively process nonregistered ultrasound image pairs, a common challenge in medical imaging, across various clinical settings and device types. By addressing these challenges, the research seeks to provide a more generalized and adaptable solution that can be widely applied across diverse medical scenarios, improving the accessibility and quality of diagnostic imaging.

Methods: A retrospective analysis was conducted by using a cycle-consistent generative adversarial network (CycleGAN) framework enhanced with perceptual loss to improve the quality of ultrasound images, focusing on nonregistered image pairs from various organ systems. The perceptual loss was integrated to preserve anatomical integrity by comparing deep features extracted from pretrained neural networks. The model's performance was evaluated against corresponding high-resolution images, ensuring that the enhanced outputs closely mimic those from high-end ultrasound devices. The model was trained and validated using a publicly available, diverse dataset to ensure robustness and generalizability across different imaging scenarios.

Results: The advanced CycleGAN framework, enhanced with perceptual loss, significantly outperformed the previous state-of-the-art, stable CycleGAN, in multiple evaluation metrics. Specifically, our method achieved a structural similarity index of 0.2889 versus 0.2502 (P<.001), a peak signal-to-noise ratio of 15.8935 versus 14.9430 (P<.001), and a learned perceptual image patch similarity score of 0.4490 versus 0.5005 (P<.001). These results demonstrate the model's superior ability to enhance image quality while preserving critical anatomical details, thereby improving diagnostic usefulness.

Conclusions: This study presents a significant advancement in ultrasound imaging by leveraging a CycleGAN model enhanced with perceptual loss to bridge the quality gap between images from different devices. By processing nonregistered image pairs, the model not only enhances visual quality but also ensures the preservation of essential anatomical structures, crucial for accurate diagnosis. This approach holds the potential to democratize high-quality ultrasound imaging, making it accessible through low-cost portable devices, thereby improving health care outcomes, particularly in resource-limited settings. Future research will focus on further validation and optimization for clinical use.

Keywords: cycle generative adversarial network; generative networks; image enhancement; imaging; machine learning; perceptual loss; portable handheld devices; ultrasound images; ultrasound scans.

Figure 1

An overview of the cycle generative adversarial network model training and loss computation framework. The solid black arrows indicate the flow of data. The dashed red arrows indicate the flow of information for loss computation.

Figure 2

The model architectures. The (A) generator and (B) discriminator model architectures. The figure legend lists the different layers in the models. ReLU: rectified linear unit.

Figure 3

A comparative visualization of ultrasound scans from the test set, showcasing the performance of different enhancement frameworks on the same high-low quality image pair. Each subfigure corresponds to a different model’s output, allowing for a direct comparison of the anatomical preservation and image quality achieved by each approach. (A) reference low, (B) Pix2Pix [27], (C) MSPGAN [11], (D) RegGAN [23], (E) reference high, (F) CycleGAN [13], (G) stable CycleGAN [12], and (H) proposed method. CycleGAN: cycle-consistent generative adversarial network; MSPGAN: multilevel structure-preserved generative adversarial network; RegGAN: registration generative adversarial network.

Figure 4

A comparative visualization of ultrasound scans from the test set, showcasing the performance of different enhancement frameworks on the same high-low quality image pairs. (A) Thyroid, (B) carotid, (C) liver, (D) kidney, and (E) breast. CycleGAN: cycle-consistent generative adversarial network; MSPGAN: multilevel structure-preserved generative adversarial network; RegGAN: registration generative adversarial network.