Early detection is key to intervening with students diagnosed with a specific learning disorder (SLD), which includes problems with spelling, grammar, punctuation, and clarity and organization of written expression, as a means of preventing potential negative consequences from this disorder. Deep convolutional neural networks (CNNs) perform better than human beings in many visual tasks such as making a medical diagnosis from visual data. The purpose of this study was to evaluate the ability of a deep CNN to detect students with a diagnosis of an SLD based on their handwriting. A so-called MobileNetV2 deep CNN architecture was used by applying transfer learning. The model was trained using a data set of 497 images of handwriting samples from students with a diagnosis of an SLD as well as students without this diagnosis. The detection of an SLD on the validation set yielded a mean area under the receiver operating characteristics curve of 0.89. This novel attempt to detect students with the diagnosis of an SLD using deep learning can potentially provide fast initial screening of students who may meet the criteria for a diagnosis of an SLD, thereby improving their chances of effective intervention.