9 Introduction, aims and outline of the thesis 1 has an important role (next to endoscopy) to evaluate the local tumour response after neoadjuvant treatment. This ‘restaging’ helps determine whether patients still require routine surgical resection, or may be managed with more conservative treatment alternatives such as watch and wait (W&W) [11]. Considering that rectal cancer treatment is increasingly tailored to a patient’s individual needs, the radiologist now plays an important role as a sparring partner in the multidisciplinary team (MDT) to help determine these needs and push personalized medicine forwards. To support radiologists in fulfilling this role, various imaging guidelines, reporting guides and staging templates have been developed for rectal cancer [12-14]. In addition, new diagnostic grading systems have recently been published to assist radiologists in assessing the local tumour response after neoadjuvant therapy [15-18]. An important goal of these templates and grading systems is to enhance uniformity in radiological reporting and thereby promote consistent and evidence-based patient management. However, there are several pitfalls and challenges when it comes to the application and clinical implementation of such tools. First of all, several reports have shown that the experience level of the radiologist has a significant impact on the diagnostic performance and, consequently, on MDT decisions and treatment outcomes [19-22]. New diagnostic methods, such as response-grading systems have often only been tested by small groups of expert radiologists before being adopted into guidelines [15-18, 22]. Whether such tools are also reproducible and accurate in the hands of radiologists in everyday clinical practice is often poorly examined. An important reason for this is that it is challenging to set up studies to test and validate diagnostic methods on a large scale, i.e. with multiple radiologists and using data from different clinical centres. Furthermore, limited data is available on how well staging guides and templates, such as those published by the European Society of Gastrointestinal and Abdominal Radiology (ESGAR), are adopted in daily clinical routine, what their clinical impact is and whether there are any practical limitations that need to be addressed and investigated further. Required research infrastructure An important missing link to address the challenges described above is the availability of a user-friendly platform to enable large scale diagnostic validation studies where images from different clinical centres can easily be shared and evaluated by multiple readers. Though recent “image biobank” and “data repository” initiatives do offer platforms to share research data, these platforms are often primarily focused on the development of artificial intelligence (AI) tools [23, 24]. Digital Imaging and Communications in Medicine (DICOM)-viewers embedded within these platforms – if any – are typically very basic and mainly offer tools for AI annotation. Functionality dedicated to visual diagnostic image interpretation is limited and these platforms are therefore poorly equipped to support more clinically oriented diagnostic studies. Proper validation of visual diagnostic methods is critical, considering the direct impact on patient management outlined
RkJQdWJsaXNoZXIy MjY0ODMw