From research to clinical practice: lessons learnt from the Cerefy brain atlases

Abstract

Purpose Experience in development and commercialization of the Cerefy brain atlases is shared and compared with some other neuroimaging resources. Materials and Methods Cerefy includes four groups of atlases such as anatomical, functional, vascular and blood supply territories. This project developed 17 brain atlas products being used worldwide in clinical practice, medical research, and education. These atlases are incorporated in more than 1,000 neurosurgical workstations and distributed on over 4,000 CD-ROMs to individual clinicians and medical schools. These brain atlas products were compared with several academic and market performance indicators. Results Cerefy resulted in 13 patents filed and 3 granted, 44 journal papers, 10 international awards, and revenue from commercial products. Numerous anonymous patients have been benefited directly or indirectly from this work, but the impact on surgical practice is more difficult to assess. Conclusion The experience and the lessons learned in the Cerefy brain atlas development were identified, so this can guide other research groups in their translational efforts from bench to bedside.

Human brain atlasing: past, present and future

“…The lessons learnt from the experience in commercialisation of our brain atlases are summarised and discussed in Nowinski, 91 so this experience can guide other research groups in their translational efforts from bench to bedside. That work 91 also summarised my first 10-year brain atlas efforts, from the release of the first atlas resulting then in 17 atlas products developed, 4000 CD ROMs distributed by Thieme, nine awards from clinical societies and 13 patents filed and three granted. The comparison of the 20-year to 10-year outcomes demonstrates the dynamics of our work: a jump in intellectual property generation is especially evident.…”

Automatic processing of diffusion-weighted ischemic stroke images based on divergence measures: slice and hemisphere identification, and stroke region segmentation

“…Forty-eight data sets were from 1.5 T and nine were from 3 T machines. The data set had following types: lacunar (14), ischemic with hemorrhagic transformation (10) and non-lacunar (33). Amongst these total 57 cases, four were low SNR (1 lacunar and 3 nonlacunar) and 2 bilateral stroke cases.…”

“…We have developed a computer assisted diagnosis (CAD) system [13,14] for rapid stroke image analysis with an initial algorithm for infarct segmentation based on energy measures. To enhance the outcome, Gupta et al [15] identify stroke slice and hemisphere automatically based on parameters derived from percentiles of intensity of the image that can be a preprocessor to segmentation.…”

Robust calculation of the midsagittal plane in CT scans using the Kullback–Leibler’s measure

“…7 that the ratio of semi-major to minor axes lengths in the top-most slices increasing from 1.02 to 1.12 (slice [22][23][24], suggesting that the skull shape changes from circular to elliptical again. However, the angular discrepancy remains increasing.…”

“…This approach has been validated for MR images (T1, T2, MRA, SPGR, FLAIR) and tried initially for a few CT scans. This method is further extended for diffusion and perfusion weighted images [22], showing that it may work across different pulse sequences. These two approaches have been incorporated into our MR stroke CAD system [23] trial licensed to several hospitals and companies worldwide.…”