Inversion Time (TI) is defined as the interval between two RF pulses with net magnetization in opposite direction with respect to the static magnetic field. It is extremely relevant with MRI because it directly influences the resulting image. In particular, TI should be carefully selected in CMR with late-gadolinium enhancement: if properly chosen, pathology-free myocardium appears fully black in the final acquired image, with a high level of contrast with respect to scar tissue.
Hence, optimized TI maximizes the chances of correctly discriminating the portions of tissue with different histocellular compositions, such as portions of pathology-free tissue versus portions of tissue with an acute or chronic pathologic process. A poor selection of TI leads to images that may not support a diagnosis, or even lead to an inaccurate diagnosis. This implies extra costs, and possibly the repetition of the whole exam.
TI is usually manually estimated by the MRI operator, based on personal experience. Alternatively, several images can be acquired at different TI inversion times and subsequently evaluate to define the correct TI. This process mandates adequately trained personnel and it is expensive, both in terms of time and resources used, and the quality of the images obtained is not guaranteed.
Some months ago, a research group formed by Dr. Camilla Torlasco (Istituto Auxologico Italiano, Milan, Italy), Prof. Daniela Besozzi and Dr. Daniele Papetti (both with the University of Milano-Bicocca, Italy) and me (Ca’ Foscari) developed and patented an AI-based methodology for automatic TI estimation (codename “THAITI”, patent code IT102022000017907), later extended with international PCT application. We trained a model able to estimate the optimal TI interval for a patient, in a remarkable example of precision medicine.
Since its development, THAITI obtained two important awards: the best abstract award from the Italian Society of Cardiology (SIC), and a “Shark Tank” award during the Global CMR 2024 conference in London, UK. Both conferences appreciated THAITI’s innovative character, the translational value, and the advanced TRL. We are now finalizing the implementation and looking for investors interested in bringing THAITI on the market!
If you are interested in THAITI or need additional information feel free to contact me (marco.nobile@unive.it) or watch the PowerPoint presentation about THAITI that we presented at CMR2024.
