Ultrasound imaging techniques are widely used in clinical applications. The main component of an ultrasound machine is a beamformer. The beamformer plays a major role in the final reconstructed image quality. For a properly reconstructed image of high quality, the main lobe of the point target should be narrow and the side lobes should be reduced.
Delay and Sum (DAS) beamformer is the most commonly used technique in ultrasound imaging due to its simplicity. However, it has low image resolution and less off-axis interference rejection. Thus, efforts have been made to get a clearer image by changing the beamformer.
A non-linear beamforming technique called Filtered Delay Multiply and Sum (F-DMAS) was introduced to compensate for the drawbacks of the delay and sum beamformer. With F-DMAS beamformer, the reconstructed image resolution was clearer and several similar techniques were developed such as the double-stage DMAS and the Baseband-DMAS. However, contrast to noise ratio was limited in these methods. Recently, we had introduced two other non-linear beamforming techniques called Filtered Delay Weight Multiply and Sum (F-DwMAS) and Filtered Delay Euclidian-weighted Multiply and Sum (F-DewMAS) to push the state-of-the art by applying it to Synthetic Aperture (SA) technique, which is a technology gaining traction only recently. However, current ultrasound imaging scanners widely utilize Conventional Focused Beamforming (CFB) technique. CFB technique has the advantages of requiring less memory for data handling and has a good depth of penetration for imaging tissues deep inside. These advantages have made CFB more practical and ubiquitous. Nevertheless, CFB has a drawback when practically implementing the transmit beam pattern. The transmit beam pattern changes for different transmissions and the beam must be steered asymmetrically. This is because of insufficient number of elements when approaching the edges. Thus, it is well recognized that CFB yields good image quality only around the focal region and the operators are trained to keep the target of interest around the centre of the image while scanning.
Mr. Anudeep Vayyeti 
Prof. Arun Kumar Thittai
In this work, Mr. Anudeep Vayyeti and Prof. Arun Kumar Thittai, from the Department of Applied Mechanics, Indian Institute of Technology Madras, Chennai, India, have proposed a new beamforming technique called Filtered Delay optimally-weighted Multiply and Sum (F-DowMAS) that can overcome the transmit beam problem and can provide a clear and high quality reconstructed image throughout the width of the image.
Although the major interest was to improve the performance of the latest non-linear beamformer, it was also found that this technique could be extended to delay and sum (DAS) as well. F-DowMAS was compared with DAS and F-DMAS in terms of performance. It was found that F-DowMAS performed better than DAS and F-DMAS in terms of axial resolution, lateral resolution, contrast ratio, and contrast-to-noise ratio. The image resolution was also the best for F-DowMAS.
Prof. Chandra Sekhar Seelamantula from the Department of Electrical Engineering, Indian Institute of Science Bengaluru, Bengaluru, India, appreciated the work done by the authors by giving the following comments: “Ultrasound imaging is a widely used non-ionizing imaging modality in the clinical setting. Several attempts have been made in the recent past to improve the reconstruction image quality and lateral/axial resolution by employing sophisticated computational tools. The work by Prof. Arun Thittai and team is a state-of-the-art computational approach to improve the resolution both in the axial and lateral directions, imaging contrast, and contrast-to-noise ratio. These are objective parameters based on which the performance of an ultrasound system can be rated. Prof. Arun Thittai’s system offers significant improvements over the benchmark methods. The subjective performance is also superior and can be appreciated from the visual quality of the reconstructed images. Prof. Arun Thittai’s latest Nature Scientific Reports publication shows convincing evidence both on phantom data and real data. A superior quality image will go a long way in clinical diagnosis. The approach proposed in Prof. Arun Thittai’s paper is computational and does not alter the image acquisition setup. Hence, it has the potential to be integrated into the existing scanners.”
Article by Akshay Anantharaman
Here is the original link to the paper:
https://www.nature.com/articles/s41598-021-00741-5
