There is a growing need for imaging breasts with silicone implants, driven by augmentation procedures for cosmetic and cancer-related reconstruction purposes [1]. While the silicone implant’s integrity accounts for the main radiological concern of these patients, breast cancer detection and diagnosis remain an important need. A promising non-invasive method to diagnose breast cancer is diffusion-weighted MRI (DWI) [2], a diagnosis based on the fact that cancerous tissues exhibit lower apparent diffusion coefficient (ADCs) than normal fibroglandular tissue or benign lesions [3]. Diffusion measurements require fast imaging techniques, exhibiting reduced sensitivity to motional effects. In most cases the method of choice is spin-echo echo planar imaging (SE-EPI). Still, SE-EPI is prone to display artifacts, particularly when applied to the examination of challenging organs like silicone-implanted breasts. Here the presence of water-, silicone-, and fat-rich regions results in chemical shift artifacts and field heterogeneities, leading to spurious replicas that may confound the radiological interpretation [4]. In this work we explore the use of a DWI imaging methodology based on SPatio-temporal ENcoding (SPEN). This single-shot technique has proven as a highly robust alternative in terms of overcoming B0-inhomogeneities and heterogeneous chemical environments [5-6]. An important advantage of SPEN rests in its ability to resolve—and thereby, if needed, suppress—the contributions of different chemical sites without any changes in the basic imaging-oriented pulse sequence [7]. The purpose of this study was to explore the application of SPEN’s strategies to acquire reliable ADC maps of fibroglandular tissue in silicone implanted human breasts, whose images have been freed from the dominant silicone signal contributions.This study was approved by the Institutional Review Board of the Wolfson Medical Center (Holon, Israel) and included seven healthy volunteers with different sizes of silicone implants and of fibroglandular tissue. Axial images were acquired at 3 Tesla (T) on a Siemens TrioTIM scanner using a four-channel bilateral breast coil. The MRI protocol included a T2-weighted turbo spin-echo (TSE) followed by DWI acquisitions of twice-refocused SE-EPI and by SPEN DWI with fat suppression using d=17 ms, ∆=35 ms and n=7 nominal b-values in the range: 0-750 (s/mm²). The spatial resolutions of both SE-EPI and SPEN experiments were 2.0x2.0x2.5 mm³. The total scan durations (per slice) were 180 ms and 277 ms for DWI SPEN and SE-EPI, respectively.While SPEN’s image retrieval process does not require Fourier processing, it follows that FT along the SPEN/time acquisition axis, will yield the sample’s chemical shift spectrum [7]. In this way SPEN provides the basis for single-scan spectroscopic imaging; individual contributions arising from each chemical resolved site can then be manually isolated (or removed). Figure 1a shows a TSE multi-shot scan of a healthy volunteer including the silicone implant, the fibroglandular tissue, and the fatty tissue. Following SPEN’s spectral filtration procedure, we were able to selectively separate the fibroglandular tissue (Fig. 1c) and the silicone implant (Fig. 1d) – both from the same single-scan information. Additionally, utilized in Figure 1 was SPEN’s capability to “zoom” on a single breast, while avoiding folding. Figure 2 shows a representative scan of a healthy volunteer, presenting a TSE image (Fig. 2a), as well as anatomical b-zero images (Fig. 2b) and ADC maps (Fig. 2c) delivered by DWI SE-EPI and by SPEN, respectively. To avoid folding artifacts the FOV used in the SE-EPI was chosen to cover the whole body, whereas in the folding-free SPEN scans the FOV could be reduced to half this size. b-zero images afforded by SE-EPI show a noticeable chemical shift displacement artifact of the silicone implant, which together with a sizable ghost of both the water and silicone signals (yellow arrows in Fig. 2b) appear below the fibroglandular tissue. Under similar conditions SPEN succeeds to deliver images that are significantly less influenced by ghosts, showing the fibroglandular tissue in its right location and a considerably weaker silicone contribution. Analyses of diffusion values measured in breast ROIs (red circled ROIs indicated in Fig. 2c), revealed no significant differences between the ADCs derived from SE-EPI and from SPEN (SE-EPI ADC =1.57±0.18 x10^-3 mm²/s, SPEN ADC =1.56±0.12 x10^-3 mm²/s, tailed p = 0.16).This study explored the use of SPEN–based strategies incorporating ADC measurements in the presence of silicone implant. This includes single-scan images that are less affected by field inhomogeneities than SE-EPI’s counterparts, as well as SPEN’s built-in capacity to separate multiple spectral contributions without demanding additions or modifications to the original single-scan 2D imaging sequence, thereby opening new screening possibilities for the identification of malignancies in breast augmented patients.EU, Kimmel Institute and Perlman Family Foundation.