A tomographic image prediction device 1 comprises an input unit 11, a prediction unit 12, an output unit 13, and a learning unit 14. The tomographic image prediction device 1: receives input of a tomographic image of the brain of a subject as an input image I1; predicts, using a deep neural network of the prediction unit 12, a tomographic image of the brain of the subject at a point in time later than when said tomographic image was acquired; and outputs the predicted tomographic image as an output image I2. The tomographic image prediction device 1 can cause the deep neural network to learn using a tomographic image database 15. Due to this configuration, a tomographic image prediction device and a tomographic image prediction method that are useful for planning a future course of treatment or the like for the subject are attained.本發明之斷層圖像預測裝置1具備輸入部11、預測部12、輸出部13及學習部14。斷層圖像預測裝置1將受試者之腦之斷層圖像輸入作為輸入圖像I1,且藉由預測部12之深度神經網路預測較該斷層圖像取得時更未來之該受試者之腦之斷層圖像,並將該預測之斷層圖像作為輸出圖像I2輸出。斷層圖像預測裝置1可使用斷層圖像資料庫15使深度神經網路學習。藉此實現有助於制訂受試者將來之治療等方針的斷層圖像預測裝置及斷層圖像預測方法。