Deciphering Doctor's Handwriting using Deep Learning

We built a bot that automatically reads doctors handwriting on Belgian death certificates with an accuracy of 47% of certificates of the usable data set correctly predicted. The bot supports government officials with the official death registration and allows for a faster such registration. The solution consists of three main components: an image-processing module, a neural net, and a natural-language processing module to output predictions of medical terms. When a person deceases, a medical practitioner must certify the deceased state of the person. There is a standard form which the physician fills in. This is done "in the field," through a handwritten statement on the form, which is subsequently forwarded to other officials under sealed envelope. The physician officially records the direct cause of death, and, if known, any secondary causes. An example of such a death certificate is given below. The example above is fairly straightforward to read, but notoriously difficult-to-read examples also exist. Because the physicians record death causes "in the field," a 100% digital solution is not feasible, and a lot of handwriting will exist for years to come. The raw data are one-page scans, provided as a PDF.