The objective of this study is to measure T2* relaxation time in the triangular fibrocartilage (TFC) disc in asymptomatic volunteers and evaluate its variation with factors such as age, hand dominance, sex, and ulnar variance, using a dedicated MRI sequence. The MRI protocol included anatomical sequences as well as a 3D ultra-short echo time (UTE)-T2* mapping sequence. A linear regression model was used to assess the potential influence of age, sex, and hand dominance on T2* values measured in the TFC disc and to evaluate the correlation between T2* values and ulnar variance. T2* relaxation time was positively correlated with age. The higher T2* relaxation times may reflect early degeneration of the fibrocartilage microstructure, which is associated with both biomechanical factors and the aging process. However, T2* was not significantly influenced by sex or hand dominance, nor was it correlated with ulnar variance (this later being limited by the fact that none of our subject had positive ulnar variance). In conclusion, UTE-T2* is a promising MRI technique showing positive correlation with age in the TFC of healthy subjects. These findings are a first step to establish normative T2* values and will help interpreting deviations observed in patient with suspected pathology in future studies.

Although minimally invasive surgery has significantly reduced the incidence of incisional hernia (IH) in colorectal procedures, the choice of specimen extraction site continues to pose a risk. This study explores how the location of extraction site influences the occurrence of IH in patients undergoing minimally invasive colorectal resections. We conducted a retrospective observational cohort study involving consecutive patients who underwent minimally invasive colorectal resection for colorectal cancer from 2013 to 2021. Patients with open surgery, previous hernia repair, reoperations during the study period or without imaging follow-up were excluded. The primary outcome measured was the CT-proven incidence of IH at the extraction site. A total of 191 patients were analyzed. Among them, 113 patients (59.2

Artificial intelligence (AI) has introduced numerous opportunities for human assistance and task automation in medicine. However, it suffers from poor generalization in the presence of shifts in the data distribution. In the context of AI-based computed tomography (CT) analysis, significant data distribution shifts can be caused by changes in scanner manufacturer, reconstruction technique or dose. AI harmonization techniques can address this problem by reducing distribution shifts caused by various acquisition settings. This paper presents an open-source benchmark dataset containing CT scans of an anthropomorphic phantom acquired with various scanners and settings, which purpose is to foster the development of AI harmonization techniques. Using a phantom allows fixing variations attributed to inter- and intra-patient variations. The dataset includes 1378 image series acquired with 13 scanners from 4 manufacturers across 8 institutions using a harmonized protocol as well as several acquisition doses. Additionally, we present a methodology, baseline results and open-source code to assess image- and feature-level stability and liver tissue classification, promoting the development of AI harmonization strategies.

Midtarsal injuries are often missed at initial presentation which may lead to long-term complications. Nonetheless, radiographs (XR) are used as a primary imaging method. The place of cone beam computer tomography (CBCT) remains unclear in the management of midfoot injuries. The aim of this study was to update imaging findings on traumatic ankle and foot injuries (TAAFI) with CBCT and to assess the sensitivity, specificity and accuracy of XR compared to CBCT for midfoot injuries detections. All CBCT studies performed due to (TAAFI) that had previous XR were collected for a period of 5 years. They were retrospectively anonymized and analyzed by a radiologist. A second blinded study of XR was made by a second radiologist as a control. A total of 754 cases were included. Lisfranc and Chopart injuries were detected in 153 (20.2

Background Artificial intelligence (AI) seems promising in diagnosing pneumonia on chest x-rays (CXR), but deep learning (DL) algorithms have primarily been compared with radiologists, whose diagnosis can be not completely accurate. Therefore, we evaluated the accuracy of DL in diagnosing pneumonia on CXR using a more robust reference diagnosis. Methods We trained a DL convolutional neural network model to diagnose pneumonia and evaluated its accuracy in two prospective pneumonia cohorts including 430 patients, for whom the reference diagnosis was determined a posteriori by a multidisciplinary expert panel using multimodal data. The performance of the DL model was compared with that of senior radiologists and emergency physicians reviewing CXRs and that of radiologists reviewing computed tomography (CT) performed concomitantly. Results Radiologists and DL showed a similar accuracy on CXR for both cohorts ( p ≥ 0.269): cohort 1, radiologist 1 75.5% (95% confidence interval 69.1–80.9), radiologist 2 71.0% (64.4–76.8), DL 71.0% (64.4–76.8); cohort 2, radiologist 70.9% (64.7–76.4), DL 72.6% (66.5–78.0). The accuracy of radiologists and DL was significantly higher ( p ≤ 0.022) than that of emergency physicians (cohort 1 64.0% [57.1–70.3], cohort 2 63.0% [55.6–69.0]). Accuracy was significantly higher for CT (cohort 1 79.0% [72.8–84.1], cohort 2 89.6% [84.9–92.9]) than for CXR readers including radiologists, clinicians, and DL (all p -values < 0.001). Conclusions When compared with a robust reference diagnosis, the performance of AI models to identify pneumonia on CXRs was inferior than previously reported but similar to that of radiologists and better than that of emergency physicians. Relevance statement The clinical relevance of AI models for pneumonia diagnosis may have been overestimated. AI models should be benchmarked against robust reference multimodal diagnosis to avoid overestimating its performance. Trial registration NCT02467192 , and NCT01574066 . Key point • We evaluated an openly-access convolutional neural network (CNN) model to diagnose pneumonia on CXRs. • CNN was validated against a strong multimodal reference diagnosis. • In our study, the CNN performance (area under the receiver operating characteristics curve 0.74) was lower than that previously reported when validated against radiologists’ diagnosis (0.99 in a recent meta-analysis). • The CNN performance was significantly higher than emergency physicians’ ( p ≤ 0.022) and comparable to that of board-certified radiologists ( p ≥ 0.269). Graphical Abstract

Incidents involving ionizing radiation pose a risk of immediate and long-term clinical consequences for both victims and responders in the event of secondary contamination. Rapid identification of the problem and a coordinated response are crucial. This article summarizes the key challenges related to the emergency management of a single patient or multiple victims, addressing the importance of recognizing such a case, radioprotection measures, decontamination, and available treatments.

To determine prevalence in the symptomatic population of dorsal mucoid cysts centered on dorsal capsuloscapholunate septum (DCSS) using high-field magnetic resonance imaging (MRI) for anatomoclinical and epidemiological correlations. This single-center retrospective study analyzed all 3-Tesla MRIs consecutively performed for painful wrists in 295 patients. Two blinded readers performed measurements. The protocol included T1 spin echo and 3D proton density sequences with fat saturation. Inter-observer reliability was assessed using kappa and intra-class correlation coefficients for cyst detection and volumetry, respectively. Disagreements concerning cyst detection were resolved by a consensus reading. Cyst size, relationship to extrinsic and scapholunate ligaments (SL), continuity of SL, minimum distance to the posterior interosseous nerve (PIN), cyst communication with joint, and anatomical classifications of cysts were analyzed. Correlation tests were performed to assess associations. Two-hundred ninety-five patients (mean age 39.6 +/− 15.6 (standard deviation), 161 males) were evaluated for detection of dorsal wrist cysts identified in 150/295. In this subgroup, the mean age was 38.7 years (15–75), the sex ratio of 0.6 (59

Background/Objectives: T1 relaxation time has been shown to be valuable in detecting and characterizing tumors in various organs. This study aims to determine whether native T1 relaxation time can serve as a useful tool in distinguishing sarcomas from benign tumors. Methods: In this retrospective study, patients with histologically confirmed soft tissue sarcomas and benign tumors were included. Only patients who had not undergone prior treatment or surgery and whose magnetic resonance imaging (MRI) included native T1 mapping were considered. Images were acquired using both 1.5 T and 3 T MRI scanners. T1 histogram parameters were measured in regions of interest encompassing the entire tumor volume, as well as in healthy muscle tissue. Results: Out of 316 cases, 16 sarcoma cases and 9 benign tumor cases were eligible. The T1 values observed in sarcoma did not significantly differ from those in benign lesions in both 1.5 T and 3 T MRIs (p1.5T = 0.260 and p3T = 0.119). However, T1 values were found to be lower in healthy tissues compared to sarcoma at 3 T (p = 0.020), although this difference did not reach statistical significance at 1.5 T (p = 0.063). At both 1.5 T and 3 T, no significant difference between healthy muscle measured in sarcoma cases or benign tumor cases was observed (p1.5T = 0.472 and p3T = 0.226). Conclusions: T1 mapping has the potential to serve as a promising tool for differentiating sarcomas from benign tumors in baseline assessments. However, the standardization of imaging protocols and further improvements in T1 mapping techniques are necessary to fully realize its potential.

Purpose: Transducer positioning for liver ablation by magnetic resonance-guided high-intensity focused ultrasound (MRgHIFU) is challenging due to the presence of air-filled organs or bones on the beam path. This paper presents a software tool developed to optimize the positioning of a HIFU transducer dedicated to abdominal thermal therapy, to maximize the treatment's efficiency while minimizing the near-field risk. Methods: A software tool was developed to determine the theoretical optimal position (TOP) of the transducer based on the minimization of a cost function using the particle swarm optimization (PSO). After an initialization phase and a manual segmentation of the abdomen of 5 pigs, the program randomly generates particles with 2 degrees of freedom and iteratively minimizes the cost function of the particles considering 3 parameters weighted according to their criticality. New particles are generated around the best position obtained at the previous step and the process is repeated until the optimal position of the transducer is reached. MR imaging data from in vivo HIFU ablation in pig livers was used for ground truth comparison between the TOP and the experimental position (EP). Results: As compared to the manual EP, the rotation difference with the TOP was on average -3.1 ± 7.1° and the distance difference was on average -7.1 ± 5.4mm. The computational time to suggest the TOP was 20s. The software tool is modulable and demonstrated consistency and robustness when repeating the calculation and changing the initial position of the transducer.

IntroductionThe trapeziometacarpal (TMC) joint is a complex joint, whose anatomy and function are different from the metacarpophalangeal joints of the long fingers. The stability of this joint is ensured at three levels by multiple structures: osteochondral, capsulo-ligamentous, and musculo-tendinous. The anatomical and biomechanical structures ensuring the stability of the TMC joint are perfectly evaluated on magnetic resonance imaging (MRI), with a high degree of confidence. All described ligaments are anatomically visible and perfectly assessed on MRI and ultrasound (US): the dorsoradial ligament, the posterior oblique ligament, the intermetacarpal ligament, the ulnar collateral ligament, the two bundles of the anterior oblique ligament (break ligament), as well as the superficial anterior oblique and deep anterior oblique ligaments.MethodologyThis educational review assesses the TMC joint anatomy using high-field MRI and US compared with cadaveric specimens as well the biomechanics of this joint. In addition, it highlights pathological patterns of traumatic (sprain, dislocation, and fractures) and degenerative diseases.Results and conclusionKnowledge of TMC joint anatomy is crucial to the radiologists’ understanding and assessment of various traumatic and degenerative pathologies, and thus helps clinicians and surgeons choose the appropriate treatment.

The aim of this study is to propose diagnostic reference levels (DRLs) values for mammography in Switzerland. For the data collection, a survey was conducted among a sufficient number of centres, including five University hospitals, several cantonal hospitals, and large private clinics, covering all linguistic regions of Switzerland to be representative of the clinical practice. The data gathered contained the mean glandular dose (MGD), the compressed breast thickness (CBT), the mammography model and the examination parameters for each acquisition. The data collected was sorted into the following categories: 2D or digital breast tomosynthesis (DBT) examination, craniocaudal (CC) or mediolateral oblique (MLO) projection, and eight categories of CBT ranging from 20 mm to 100 mm in 10 mm intervals. A total of 24 762 acquisitions were gathered in 31 centres on 36 mammography units from six manufacturers. The analysis showed that the data reflects the practice in Switzerland. The results revealed that the MGD is larger for DBT than for 2D acquisitions for the same CBT. From 20-30 mm to 90-100 mm of CBT, the 75th percentile of the MGD values obtained increased from 0.81 mGy to 2.55 mGy for 2D CC acquisitions, from 0.83 mGy to 2.96 mGy for 2D MLO acquisitions, from 1.22 mGy to 3.66 mGy for DBT CC acquisitions and from 1.33 mGy to 4.04 mGy for DBT MLO acquisitions. The results of the survey allow us to propose Swiss DRLs for mammography according to the examination type (2D/DBT), projection (CC/MLO) and CBT. The proposed values are very satisfactory in comparison with other studies.

Purpose: The purpose of this study was to analyze the prevalence of and complications resulting from temporal bone fractures in adult and pediatric patients evaluated for cranio-facial trauma in an emergency setting. Methods: A retrospective blinded analysis of CT scans of a series of 294 consecutive adult and pediatric patients with cranio-facial trauma investigated in the emergency setting was conducted. Findings were compared between the two populations. Preliminary reports made by on-call residents were compared with the retrospective analysis, which was performed in consensus by two experienced readers and served as reference standard. Results: CT revealed 126 fractures in 116/294 (39.5%) patients, although fractures were clinically suspected only in 70/294 (23.8%); p < 0.05. Fractures were longitudinal, transverse and mixed in 69.5%, 10.3% and 19.8% of cases, respectively. Most fractures were otic-sparing fractures (95.2%). Involvement of the external auditory canal, ossicular chain and the osseous structures surrounding the facial nerve was present in 72.2%, 8.7% and 6.3% of cases, respectively. Temporal bone fractures extended into the venous sinuses/jugular foramen and carotid canal in 18.3% and 17.5% of cases, respectively. Vascular injuries (carotid dissection and venous thrombosis) were more common in children than in adults (13.6% versus 5.3%); however, the observed difference did not reach statistical significance. 79.5% of patients with temporal bone fractures had both brain injuries and fractures of the facial bones and cranial vault. Brain injuries were more common in adults (90.4%) than in children (63.6%), p = 0.001. Although on-call residents reliably detected temporal bone fractures (sensitivity = 92.8%), they often missed trauma-associated ossicular dislocation (sensitivity = 27.3%). Conclusions: Temporal bone fractures and related complications are common in patients with cranio-facial trauma and need to be thoroughly looked for; the pattern of associated injuries is slightly different in children and in adults.

Purpose: We aimed to evaluate whether virtual non-contrast cerebral computed tomography (VNCCT) reconstructed from intravenous contrast-enhanced dual-energy CT (iv-DECT) could replace non-contrast CT (NCCT) in patients with suspected acute cerebral ischemia. Method: This retrospective study included all consecutive patients in whom NCCT followed by iv-DECT were performed for suspected acute ischemia in our emergency department over a 1-month period. The Alberta Stroke Program Early CT Score (ASPECTS) was used to determine signs of acute ischemia in the anterior and posterior circulation, the presence of hemorrhage, and alternative findings, which were randomly evaluated via the consensus reading of NCCT and VNCCT by two readers blinded to the final diagnosis. An intraclass correlation between VNCCT and NCCT was calculated for the ASPECTS values. Both techniques were evaluated for their ability to detect ischemic lesions (ASPECTS <10) when compared with the final discharge diagnosis (reference standard). Results: Overall, 148 patients (80 men, mean age 64 years) were included, of whom 46 (30%) presented with acute ischemia, 6 (4%) presented with intracerebral hemorrhage, 11 (7%) had an alternative diagnosis, and 85 (59%) had no pathological findings. The intraclass correlation coefficients of the two modalities were 0.97 (0.96–0.98) for the anterior circulation and 0.77 (0.69–0.83) for the posterior circulation. The VNCCT’s sensitivity for detecting acute ischemia was higher (41%, 19/46) than that of NCCT (33%, 15/46). Specificity was similar between the two techniques, at 94% (97/103) and 98% (101/103), respectively. Conclusions: Our results show that VNCCT achieved a similar diagnostic performance as NCCT and could, thus, replace NCCT in assessing patients with suspected acute cerebral ischemia.

The emergence of 7T clinical MRI technology has sparked our interest in its ability to discern the complex structures of the hand. Our primary objective was to assess the sensory and motor nerve structures of the hand, specifically nerves and Pacinian corpuscles, with the dual purpose of aiding diagnostic endeavors and supporting reconstructive surgical procedures. Ethical approval was obtained to carry out 7T MRI scans on a cohort of volunteers. Four volunteers assumed a prone position, with their hands (N = 8) positioned in a “superman” posture. To immobilize and maintain the hand in a strictly horizontal position, it was affixed to a plastic plate. Passive B0 shimming was implemented. Once high-resolution 3D images had been acquired using a multi-transmit head coil, advanced post-processing techniques were used to meticulously delineate the nerve fiber networks and mechanoreceptors. Across all participants, digital nerves were consistently located on the phalanges area, on average, between 2.5 and 3.5 mm beneath the skin, except within flexion folds where the nerve was approximately 1.8 mm from the surface. On the phalanges area, the mean distance from digital nerves to joints was approximately 1.5 mm. The nerves of the fingers were closer to the bone than to the surface of the skin. Furthermore, Pacinian corpuscles exhibited a notable clustering primarily within the metacarpal zone, situated on the palmar aspect. Our study yielded promising results, successfully reconstructing and meticulously describing the anatomy of nerve fibers spanning from the carpus to the digital nerve division, alongside the identification of Pacinian corpuscles, in four healthy volunteers (eight hands).

Incidents involving ionizing radiation pose a risk of immediate and long-term clinical consequences for both victims and responders in the event of secondary contamination. Rapid identification of the problem and a coordinated response are crucial. This article summarizes the key challenges related to the emergency management of a single patient or multiple victims, addressing the importance of recognizing such a case, radioprotection measures, decontamination, and available treatments.

Objectives: The aim of this study was to evaluate the diagnostic performance, image quality, and inter- and intra-observer agreement of the 3D T1 multi-echo fast field echo (mFFE) sequence in cervico-thoraco-lumbar vertebral fractures compared with conventional computed tomography (CT) as the gold standard. Methods: We conducted a prospective single-centre study including 29 patients who underwent spinal magnetic resonance imaging (MRI) at the surgeon’s request, in addition to CT for vertebral fracture assessment and classification. A 3D T1 mFFE sequence was added to the standard MRI protocol. Consecutively, two readers analyzed the 3D mFFE sequence alone, the 3D mFFE sequence with the entire MRI protocol, including the STIR and T1 sequences, and, finally, the CT images in random order and 1 month apart. A standardized assessment was performed to determine the presence or absence of a fracture, its location, its classification according to the Genant and AO classifications for traumatic and osteoporotic fractures, respectively, the loss of height of the anterior and posterior walls of the vertebral body, and the presence of concomitant disco-ligamentous lesions. Contingency tables, intraclass correlation coefficients, and Cohen’s kappa tests were used for statistical analysis. Results: A total of 25 fractures were recorded (48% cervical, 20% thoracic, and 32% lumbar), of which 52% were classified A, according to the AO classification system. The quality of the 3D mFFE image was good or excellent in 72% of cases. Inter-observer agreement was near perfect (0.81–1) for vertebral body height and for AO and Genant classifications for all modalities. Intra-observer agreement was strong-to-near perfect between CT and the 3D mFFE sequence. Regarding the diagnostic performance of the 3D mFFE sequence, the sensitivity was 0.9200 and 0.9600, the specificity was 0.9843 and 0.9895, and the accuracy was 0.9861 and 0.9769 for Readers 1 and 2, respectively. In addition, up to 40% of intervertebral disc lesions and 33% of ligamentous lesions were detected by the 3D mFFE sequence compared to CT, allowing four AO type A fractures to be reclassified as type B. Conclusions: The 3D mFFE sequence allows accurate diagnosis of vertebral fractures, with superiority over CT in detecting disco-ligamentous lesions and a more precise classification of fractures, which can prompt clinicians to adapt their management despite an image quality that still requires improvement in some cases. Key points: Vertebral fractures and disco-ligamentous lesions can be assessed using CT-like MRI sequences, with 3D T1 mFFE being superior to CT for the detection of disco-ligamentous lesions. CT-like images using the 3D T1 mFFE sequence improve the diagnostic accuracy of bone structures in MRI.

Background: Small bowel obstruction (SBO) is a common hospital admission diagnosis. Identification of patients who will require a surgical resection because of a nonviable small bowel remains a challenge. Through a prospective cohort study, the authors aimed to validate risk factors and scores for intestinal resection, and to develop a practical clinical score designed to guide surgical versus conservative management. Patients and Methods: All patients admitted for an acute SBO between 2004 and 2016 in the center were included. Patients were divided in three categories depending on the management: conservative, surgical with bowel resection, and surgical without bowel resection. The outcome variable was small bowel necrosis. Logistic regression models were used to identify the best predictors. Results: Seven hundred and thirteen patients were included in this study, 492 in the development cohort and 221 in the validation cohort. Sixty-seven percent had surgery, of which 21% had small bowel resection. Thirty-three percent were treated conservatively. Eight variables were identified with a strong association with small bowel resection: age 70 years of age and above, first episode of SBO, no bowel movement for greater than or equal to 3 days, abdominal guarding, C-reactive protein greater than or equal to 50, and three abdominal computer tomography scanner signs: small bowel transition point, lack of small bowel contrast enhancement, and the presence of greater than 500 ml of intra-abdominal fluid. Sensitivity and specificity of this score were 65 and 88%, respectively, and the area under the curve was 0.84 (95% CI: 0.80–0.89). Conclusion: The authors developed and validated a practical clinical severity score designed to tailor management of patients presenting with an SBO.

Purpose This article evaluates the feasibility, safety, and efficacy of MRI-guided lumbar or sacral nerve root infiltration for chronic back pain. We compared the outcomes of our MRI-guided infiltrations with data from CT-guided infiltrations reported in the literature and explored the potential advantages of MRI guidance. Method Forty-eight MRI-guided nerve root infiltrations were performed using a 3 T MRI machine. The optimal needle path was determined using breathhold T2-weighted sequences, and the needle was advanced under interleaved guidance based on breathhold PD-weighted images. Pain levels were assessed using a numeric rating scale (NRS) before the procedure and up to 5 months after, during follow-up. Procedure success was evaluated by comparing patients’ pain levels before and after the infiltration. Results The MRI-guided infiltrations yielded pain reduction 1 week after the infiltration in 92% of cases, with an average NRS substantial change of 3.9 points. Pain reduction persisted after 5 months for 51% of procedures. No procedure-related complications occurred. The use of a 22G needle and reconstructed subtraction images from T2 FatSat sequences improved the workflow. Conclusion Our study showed that MRI-guided nerve root infiltration is a feasible, safe, and effective treatment option for chronic back pain. Precise positioning of the needle tip and accurate distribution of the injected solution contributed to the effectiveness of MRI-guided infiltration, which appeared to be as accurate as CT-guided procedures. Further research is needed to explore the potential benefits of metal artifact reduction sequences to optimize chronic back pain management.

MRI T1-mapping is an important non-invasive tool for renal diagnosis. Previous work shows that ΔT1 (cortex-medullary difference in T1) has significant correlation with interstitial fibrosis in chronic kidney disease (CKD) allograft patients. However, measuring cortico-medullary values by manually drawing ROIs over cortex and medulla (a gold standard method) is challenging, time-consuming, subjective and requires human training. Moreover, such subjective ROI placement may also affect the work reproducibility. This work proposes a deep learning-based 2D U-Net (RCM U-Net) to auto-segment the renal cortex and medulla of CKD allograft kidney T1 maps. Furthermore, this study presents a correlation of automatically measured ΔT1 values with eGFR and percentage fibrosis in allograft kidneys. Also, the RCM U-Net correlation results are compared with the manual ROI correlation analysis. The RCM U-Net has been trained and validated on T1 maps from 40 patients (n = 2400 augmented images) and tested on 10 patients (n = 600 augmented images). The RCM U-Net segmentation results are compared with the standard VGG16, VGG19, ResNet34 and ResNet50 networks with U-Net as backbone. For clinical validation of the RCM U-Net segmentation, another set of 114 allograft kidneys patient’s cortex and medulla were automatically segmented to measure the ΔT1 values and correlated with eGFR and fibrosis. Overall, the RCM U-Net showed 50% less Mean Absolute Error (MAE), 16% better Dice Coefficient (DC) score and 12% improved results in terms of Sensitivity (SE) over conventional CNNs (i.e. VGG16, VGG19, ResNet34 and ResNet50) while the Specificity (SP) and Accuracy (ACC) did not show significant improvement (i.e. 0.5% improvement) for both cortex and medulla segmentation. For eGFR and fibrosis assessment, the proposed RCM U-Net correlation coefficient (r) and R-square (R2) was better correlated (r = -0.2, R2 = 0.041 with p = 0.039) to eGFR than manual ROI values (r = -0.19, R2 = 0.037 with p = 0.051). Similarly, the proposed RCM U-Net had noticeably better r and R2 values (r = 0.25, R2 = 0.065 with p = 0.007) for the correlation with the renal percentage fibrosis than the Manual ROI results (r = 0.3, R2 = 0.091 and p = 0.0013). Using a linear mixed model, T1 was significantly higher in the medulla than in the cortex (p<0.0001) and significantly lower in patients with cellular rejection when compared to both patients without rejection and those with humoral rejection (p<0.001). There was no significant difference in T1 between patients with and without humoral rejection (p = 0.43), nor between the types of T1 measurements (Gold standard manual versus automated RCM U-Net) (p = 0.7). The cortico-medullary area ratio measured by the RCM U-Net was significantly increased in case of cellular rejection by comparison to humoral rejection (1.6 +/- 0.39 versus 0.99 +/- 0.32, p = 0.019). In conclusion, the proposed RCM U-Net provides more robust auto-segmented cortex and medulla than the other standard CNNs allowing a good correlation of ΔT1 with eGFR and fibrosis as reported in literature as well as the differentiation of cellular and humoral transplant rejection. Therefore, the proposed approach is a promising alternative to the gold standard manual ROI method to measure T1 values without user interaction, which helps to reduce analysis time and improves reproducibility.