ABSTR AC TPurpose Based on evaluation and examination results of students, a necessity for improvement of so far purely instructorbased radiological teaching at the local institution was determined. Aim of our study was to use one out of eight seminars to exemplify adaptation of the teaching concept according to learning theory knowledge, to determine the resulting effects and to interpret them. Materials and methodsThe institutional review board approved the prospective study of the seminar conversion, which was performed after the end of the winter semester 2015/2016. Didactically, this included a course split into online preparation, attendance phase and online follow-up with integration of interactive scaffolding, practice-oriented clinical teaching according to Stanford, Peyton skills transfer and extensive feedback into the attendance phase. At the beginning and at the end of each course, each student filled in identical, standardized questionnaires (n = 256 before and after conversion) using a 5-point Likert scale (1: very good; to 5: deficient) and additionally answered two randomly chosen written examination questions from a content-adapted questionnaire pool of the last five years. For statistical evaluation, the Mann-Whitney U-Test was used for evaluation data and Fisher's Exact test for exam questions.Results Before/after conversion, the subjective total evaluation score of students was 3.22 (mean value) ± 1.51 (standard deviation) / 1.66 ± 0.78 (p < 0.001) and the objective proportion of correctly answered examination questions in the respective cohort at the beginning of the seminar 37.7/ 53.9 % and at the end of the seminar 55.1/84.6 % (p < 0.001). ConclusionThe conversion of the test seminar resulted in both a better evaluation of the teaching unit by the students (evaluation) and a considerably higher rate of correctly answered examination questions from past state examinations (learning success). This supports transferring the concept to comparable teaching units. Ergebnisse Vor/nach Umstellung betrug die subjektive studentische Gesamtbewertung 3,22 (Mittelwert) ± 1,51 (Standardabweichung) / 1,66 ± 0,78 (p < 0,001) und der objektive Anteil korrekt beantworteter Prüfungsfragen in der jeweiligen Kohorte zu Beginn des Seminars 37,7/53,9 % sowie am Ende des Seminars 55,1/84,6 % (p jeweils < 0,001).