Cutting edge of medical imaging

To be able to see what lies beneath is a virtue. We know how useful scans such as ultrasound, CT and PET are. However, to see what is happening at a deeper, molecular level, we need far more sophisticated techniques. This ability, in turn, gives scientists clues to which drugs may work. Imaging techniques, therefore, play a big role in drug discovery and diagnostics.

There have been rapid advancements in imaging techniques. Even as we become familiar with the application of techniques such as Raman spectroscopy and mass spectrometry in biology, newer ones are emerging. Two of these are the cryogenic electron microscopy (cryo-EM) and photoacoustic imaging.

India seems to be getting up to speed in both. Apart from adding muscle to drug and medical research, imaging technologies also offer immense hardware manufacturing opportunities to Indian industry.

Recently, the Science Engineering and Research Board (SERB), the government’s science research funding agency, announced that India would get four more cryo-EMs — the two existing ones are at IISc, Bengaluru, and RCB, Faridabad. The country, meanwhile, needs “at least ten more”, according to R Gopalan, a Fellow of the Electron Microscopy Society of India and the Regional Director of the Hyderabad-based International Advanced Research Centre for Powder Metallurgy and New Materials (ARCI).

However, four is a good start, and these are to come up at the IITs of Madras, Bombay and Kanpur and the Bose Institute, Kolkata.

What is cryo-EM?

It is a protein imaging technique. Proteins, as we know, are long chains of amino acids, which are themselves compounds of nitrogen, carbon, hydrogen and oxygen. There are hundreds of thousands of proteins. A protein can be a friend (bone and muscle builder) or an enemy (snake venom and the ‘spike proteins’, which are the attacking ‘arms’ of a coronavirus). To know the structure of a protein is of paramount importance.

The conventional method, called X-ray crystallography, has been to make crystals of proteins, bombard them with X-rays and understand the structure by reading the bounce-back. But growing protein crystals is time-consuming (may take months) and not always possible.

Along came cryo-EM, which fetched the 2017 Nobel prize for the three scientists who developed it. The principle is to simply flash-freeze the proteins into a solid and then bombard them with electrons. The freezing is necessary as, otherwise, the speeding electrons may push and alter the structure of the protein; or they may even get absorbed by the proteins rather than bouncing back into the detector.

Thalappil Pradeep, professor of Chemistry at IIT Madras, tells Quantum that with cryo-EMs, India can be quicker in its response to any new disease, but there is more to do with the machines. “We can make frontier discoveries,” he says of its use in pure science research. As for industrial applications other than drug discovery, there are tremendous opportunities.

Gopalan points out that cryo-EMs provide high-resolution images of molecular structures and are emerging as a “major tool” globally. The SERB is keen to have many cryo-EMs in India, he told Quantum .

It is understood that the four machines are likely to become operational by the end of 2022. IIT Madras will focus on bio-interface (between cells and microbes); IIT Bombay on ribosome translation and its implication in disease, antibiotic resistance and neurodegenerative disorders; IIT Kanpur on macromolecular structures and drug discovery with specific focus on membrane proteins; and the Bose Institute on the structure-guided drug discovery and therapeutics research.

Recently, the National Institute of Technology Andhra Pradesh (NIT-AP) at Tadepalligudem, West Godavari district, and Pennsylvania State University (PSU), USA, announced that they had jointly built an artificial intelligence-augmented photoacoustic imaging technique for cancer diagnostics. “This technique has been gaining ground in recent years by enabling precise and early-stage diagnosis of cancer, neurological disorders and vascular diseases,” says Sri Rajasekhar Kothapalli of the Department of Biomedical Engineering, PSU.

This technology, like Raman spectroscopy, shoots a beam of light (laser pulse) at the sample. Whereas in Raman spectroscopy you analyse the wavelengths of the reflected light to divine what the sample is made of, photoacoustic imaging analyses sound. This is somewhat like ultrasound scans, where you send ultrasound waves and analyse the echoes.

When a laser pulse is thrown at a tissue, the coloured molecules that are either naturally present in the body (such as haemoglobin) or introduced into the body (dyes) absorb a part of the energy and expand (thermo-elastic expansion). This expansion generates high-energy sound waves. Acoustic detectors pick up these ultrasound signals and an image is digitally reconstructed.

Each molecule has a “distinct light absorption profile” and the photoacoustic signal they emit is commensurate with the light absorbed. “We use spectroscopic (multi-wavelength) photoacoustic imaging, followed by spectral unmixing algorithms to differentiate signals coming from different molecules,” explains Sri Phani Krishna Karri of NIT-AP, who was involved in the research.

An artificial intelligence model constructs the ‘independent component’ of each colour-giving molecule (chromophore) from the given photoacoustic data.

For an AI model to train itself to become more robust you need more and more data. But there isn’t enough clinically relevant photoacoustic data. To overcome this problem, two students of Kothapalli and Karri, Sumit Agrawal (PSU) and Prameth Gaddale (NIT-AP), developed a “simulation platform” that can generate large datasets.

Photoacoustic systems are built on the conventional ultrasound imaging platform; the only additional cost is of the light source. As such, the cost to a patient is comparable to an ultrasound scan.

The hardware comprises a light source, ultrasound detector, and data acquisition and image reconstruction systems such as the ‘field programmable gate array’, says Karri. He stresses that there is “a huge potential for Indian industry to get into the manufacture of these systems”.

We value your feedback. Do send your comments to quantum@thehindu.co.in

COMMENT NOW