Research Focus

Figure 1 and Figure 2
© dkfz.de

The breast imaging research group of the Department of Radiology at the German Cancer Research Center (DKFZ) is working on the preclinical development and clinical translational application of new/improved imaging methods for the early detection and reliable characterization of suspicious findings in the breast. Due to the high social relevance of the topic, a clinical translational aspect of the research projects is of special importance for us. In established cooperation with certified radiological breast care centers and in national and international collaborations with university hospitals, we build up a link between basic preclinical research/method development and clinical application.

The aim of the breast imaging research group is to reliably detect and characterize findings in the female breast non-invasively by means of a new imaging methods based on MRI (magnetic resonance imaging) and to evaluate and implement new approaches of post-processing and artificial intelligence, radiomics and deep learning strategies. A central focus of the research projects currently is, in particular, the translational clinical evaluation of diffusion-weighted magnetic resonance tomography (DWI-MRT) in this context (Figure 1 and Figure 2).

In addition to the evaluation of diagnostic strengths and reliability of DWI-MRI for the characterization of suspicious breast lesions, new MRI techniques representing the entire spectrum of MR technology (chemical shift exchange saturation transfer, susceptibility weighted sequences and other) are evaluated. Further research projects are evaluating optimal combinations of complementary imaging techniques for breast imaging (e.g. combination of MRI with ultrasound or x-ray mammography). With the development and translational implementation of new quality assurance measures (QM management) the evaluation of validity and repeatability of new measures as part of the clinical implementation process is a further important aspect of our research projects. Several current research projects are ongoing.

Abbreviated DWI magnetic resonance mammography as integral part of the breast cancer screening clarification process

Contact: Bickelhaupt/Schlemmer

Numerous lesions detected in X-ray breast cancer screening will eventually turn out to be benign after biopsy with the risk of causing unnecessary anxiety, invasive procedures and costs. Using abbreviated magnetic resonance protocols based on breast optimized diffusion weighted imaging (DWI) might improve breast cancer screening programs by offering a non-invasive contrast agent free method to reliably characterize suspicious lesions detect with other imaging modalities. To evaluate DWI MRM as a clarification method in breast cancer screening programs a translational prospective multicentric study has been conducted demonstrating the diagnostic potential of the method.

Evaluation of workflow related reading strategies for abbreviated breast MRI protocols including contrast enhanced and non-contrast enhanced approaches

Contact: Bickelhaupt/Schlemmer

Using novel imaging technologies and new imaging approaches might not only improve the lesion detection and/or characterization. It might also allow using advanced visualization strategies that allow to more safely identify and to more clearly demonstrate imaging findings. We therefore evaluate different visualization strategies for breast imaging reading. Amongst we could demonstrate the usability of using image fusion techniques combining tissue-morphological sequences with biophysiological sequences in abbreviated imaging protocols.

Radiomics in magnetic resonance mammography using both diffusion weighted imaging (DWI) and contrast enhanced abbreviated protocols

Contact: Bickelhaupt/Bonekamp

Radiomics have gained an increasing interest in the field of radiology. Using this quantitative and systematic imaging information might be of special importance to enhance the utility and diagnostic value of abbreviated breast imaging protocols. We therefore evaluate different abbreviated imaging strategies in combination with post-processing radiomics in order to further increase the extractable quantitative image information.

Multiparametric diffusion weighted MRI – new methods for a holistic insight in tissue microstructure

Contact: Bickelhaupt/Steudle

Diffusion Weighted Imaging (DWI) provides both the possibility for focused visual assessments and for quantitative image analysis. Using different multiparametric quantitative assessments strategies of DWI might allow to characterize tissue parameters of varying histopathological microstructural correlates. Since malignant tissue transformation changes the underlying microstructure this might allow for new personalized imaging methods in oncologic imaging both in breast imaging and other organ systems.

Clarification of suspicious lesions in pre-screening age – abbreviated DWI MRI as a problem solving strategy.

Contact: Bickelhaupt/Schlemmer

Young women with unclear clinical findings in the breast can be a challenging task for breast radiologist related both to the commonly high breast tissue density and the increasing concerns regarding radiation exposure. We therefore prospectively evaluate the potential of abbreviated contrast agent free imaging approaches based on diffusion weighted imaging for the clarification of suspicious lesions in pre-screening age in a multicentric study concept.

New methods of standardization and quality assurance in DWI with new phantoms and software

Contact: Bickelhaupt/Laun; Kuder

Diffusion weighted imaging (DWI) allows for quantitative assessments of microstructural tissue correlates. However DWI implies various options of sequence adaption and modification which makes it challenging to compare the results of measurements between timepoints, vendors, places and field strengths. To address this issue we evaluate novel techniques of using standardized protocols focused on the clinical task combined with periodical measurements of dedicated DWI phantoms over timepoints and in different imaging centers to evaluate and improve the standardization and comparability of DWI in oncologic imaging.

Chemical Exchange Saturation Transfer (CEST) in high-field breast MRI at 7T and 3T

Contact: Bickelhaupt/Paech

CEST imaging provides a novel molecular imaging MRI technique that can assess endogenous proteins. Different CEST techniques have scientifically been established in the past and might be of benefit as well in the field of breast imaging. We therefore evaluate the potential of CEST in both 7T and 3T environments for breast lesion detection and characterization.

Artificial intelligence in breast magnetic resonance imaging (MRI): CNNs and Deep Learning for reader independent assessment strategies

Contact: Bickelhaupt/Jaeger

Artificial intelligence might allow more reliably detecting and characterizing imaging findings in the future. Integrating all data processing components into an artificial intelligence (AI) architecture might help to reduce fitting instabilities and information loss related to model simplicity. This might allow for improved target-specific optimization, signal exploitation, global representation learning and clinical classification using a dedicated Convolutional Neuronal Network (CNN) approach. We therefore evaluate how these advanced computational methods can be implemented into clinical decision making in breast imaging.

Diffusion Weighted Imaging (DWI) strategies in gynecologic oncologic imaging

Contact: Bickelhaupt/Mokry

Pelvic gynecologic findings are commonly difficult to characterize using conventional imaging strategies. This leads to a relatively high number of benign findings in surgical exploration. New imaging methods combining morphologic and microstructural image information based on dedicated tissue modelling using diffusion weighted imaging might improve the accuracy of oncologic imaging of the female pelvis. We therefore examine different image strategies both for lesion characterization, therapy response and tumor staging.

Selected Publications

Bickelhaupt, S., F. B. Laun, et al. (2016). "Fast and Noninvasive Characterization of Suspicious Lesions Detected at Breast Cancer X-Ray Screening: Capability of Diffusion-weighted MR Imaging with MIPs." Radiology 278(3): 689-697.

Bickelhaupt, S., D. Paech, et al. (2017). "Prediction of malignancy by a radiomic signature from contrast agent-free diffusion MRI in suspicious breast lesions found on screening mammography." J Magn Reson Imaging 2(10): 25606.

Bickelhaupt, S., J. Tesdorff, et al. (2017). "Independent value of image fusion in unenhanced breast MRI using diffusion-weighted and morphological T2-weighted images for lesion characterization in patients with recently detected BI-RADS 4/5 x-ray mammography findings." Eur Radiol 27(2): 562-569.

Bickelhaupt, S., F. B. Laun, et al. (2017) “On a fractional order calculus model in diffusion weighted breast imaging to differentiate between malignant and benign breast lesions detected on X-ray screening mammography.” PLoS One 12(4):e0176077

Bickelhaupt S, Schlemmer HP.  Verkürztes MRT in der Brustkrebsfrüherkennung.  Dtsch Arztebl International. 2016;113:[24]

Funding

The breast imaging research group is thankful for the substantial support of different funding organisations. Competitive funding was acquired over the last years covering >600.000 € of funding from e.g. the Dietmar Hopp Foundation and the German Research Foundation.

to top