Performance of deep learning models for predicting preinvasive, minimally invasive, or invasive adenocarcinoma was improved by combining binary and ternary classification models for predicting invasiveness and adjudication of discordant classification.