Naslov
LOW-COST SOLUTION FOR AUTOMATING WHOLE SLIDE IMAGING

Autori
Vinicki, Krunoslav ; Mihoković, Ivana ; Medven Zagradišnik, Lidija ; Huber, Doroteja ; Šoštarić-Zuckermann, Ivan Conrado ; Gudan Kurilj, Andrea ; Artuković, Branka ; Grabarević, Željko ; Hohšteter, Marko

Vrsta, podvrsta i kategorija rada
Sažeci sa skupova, sažetak, stručni

Izvornik
27th Ljudevit Jurak International Symposium on Comparative Pathology with One Health Session FINAL PROGRAM & BOOK OF ABSTRACTS / Krušlin, Božo ; Belicza, Mladen ; Mikuz, Gregor ; Čupić, Hrvoje ; Tkalčić, Suzana - Zagreb : European Society of Pathology, Croatian Academy of Medical Sciences, Ministry of Science, Education and Sports of the Republic of Croatia - MZOS, Institute for Clinical Medical Research of Clinical Hospital Center Sestre Milosrdnice, Zagreb City of Zagreb - City Office for Health, 2019, 47-47

Skup
27th Ljudevit Jurak International Symposium on Comparative Pathology

Mjesto i datum
Zagreb, Hrvatska, 31.05.2019. - 01.06.2019

Vrsta sudjelovanja
Poster

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
automating imaging, low cost solution, whole slide imaging, histology

Sažetak
Whole slide imaging (WSI) has significant potential to transform pathology practice. Its main function is the digitalization of glass slides, but through that process, it will also open a door for a wide array of other digital tools and algorithms, notably for machine learning (ML) and artificial intelligence (AI). In the last few years, these methods have demonstrated impressive results in image segmentation, object classification and clinical outcomes prediction, in some cases even exceeding human-level performance. Coupled with these tools, WSI will undoubtedly enhance pathology workflows. Unfortunately, due to the cost-prohibitive nature of commercial systems, relatively few pathology departments have implemented this technology, especially in underprivileged parts of the world. So, to make this method more accessible, we created a low- cost solution by automating an existing microscope. Using Arduino board with inexpensive stepper motors and a camera we managed to automate microscope stage and capture image data with sufficient precision to successfully stitch them into a large 2D digital slide. With 20X objective, the whole process of image acquisition and stitching takes, on average, 3.5h per slide. We believe that, when coupled with open‐source image processing tools and algorithms, this system can provide a viable low‐cost solution for incorporating WSI into a wide range of research programs.

Izvorni jezik
Engleski

Znanstvena područja
Veterinarska medicina

POVEZANOST RADA