Structured work-based learning in undergraduate clinical radiology immersion experience.

BACKGROUND: Practical courses in undergraduate medical training often lack a didactic concept. Active participation and learning success largely depend on chance. This study was initiated to evaluate a novel concept of structured work-based learning (WBL) in the course of students' half-day radiology immersion experience (IE). METHODS: This prospective, single-centre cohort study included 228 third-year students of the 2019 summer semester who underwent the obligatory radiology IE at a university hospital. The course was based on a novel structured WBL concept that applied established didactic concepts including blended learning, the FAIR principles of feedback, activity, individualization, and relevance, and Peyton's four-step approach. Outcomes of equal weight were student and supervisor satisfaction with the clinical radiology IE assessed by paper-based- and online survey, respectively. Secondary outcome was achievement of intended learning outcomes assessed by means of mini clinical evaluation exercises and personal interviews. RESULTS: Satisfaction with structured WBL was high in 99.0% of students. Students' expectations were exceeded, and they felt taken seriously at the professional level. Dissatisfaction was reasoned with quality of learning videos (0.6%), little support by supervisors (0.5%), or inadequate feedback (0.6%). Supervising resident physicians rated achievement of intended learning outcomes regarding cognitive and psychomotor competences as excellent for all students. Personal interviews revealed achievement of affective competence in some students. Twelve of 16 (75.0%) supervising physicians were satisfied with focussing on intended learning outcomes and student preparation for IE. Two of 15 (13.3%) supervisors were unsatisfied with time spent, and 4 of 16 (25%) with the approach of assessment. CONCLUSIONS: This study demonstrated that both students and supervisors were satisfied with the novel concept of structured WBL within the scope of clinical radiology IE. Achievement of intended learning outcomes was promising.

A pediatric prognostic score for patients undergoing allogeneic hematopoietic stem cell transplantation.

PURPOSE: The purpose of this study was to find and evaluate risk factors influencing the outcome of allogeneic hematopoietic stem cell transplantation (HSCT) in children and to develop a score stratifying patients by their risk of mortality. METHODS: We retrospectively analyzed the effects of patient and donor characteristics as well as laboratory data including liver, iron, and inflammation parameters on the overall survival of 131 children and young adults with malignant and non-malignant diseases undergoing allogeneic HSCT. RESULTS: In univariate analysis, 5-yr overall survival decreased significantly in patients with high disease risk (38% vs. 74%, P < 0.001), peripheral blood as graft source (47% vs. 73% for bone marrow, P < 0.001), ferritin >1500 ng/mL (41% vs. 79%, P = 0.001), C-reactive protein >10 mg/L (55% vs. 69%, P = 0.019), lactate dehydrogenase >6 µm∙s (22% vs. 67%, P = 0.001), gamma-glutamyl transpeptidase >1 µm∙s (43% vs. 68%, P = 0.035), and cholinesterase <60 µm∙s (36% vs. 70%, P = 0.002). For HLA recipient-donor match, there was a 5-yr overall survival of 81% for matched related, 58% for matched unrelated, 56% for mismatched unrelated, and 50% for haploidentical related donors (P = 0.020). We subsequently developed a score of parameters significant in multivariate analysis, that is, disease risk (HR = 4.1, P = 0.027), ferritin (HR = 6.4, P = 0.002), and cholinesterase (HR = 5.3, P = 0.027). For this score, 5-yr overall survival was 92% for the low-risk group, 66% for the intermediate-risk group, and 17% for the high-risk group (P < 0.001). CONCLUSIONS: Disease risk, ferritin, and cholinesterase are factors decisively influencing the prognosis after HSCT and should be evaluated in further trials.

Deep Learning CT Image Reconstruction in Clinical Practice.

BACKGROUND: Computed tomography (CT) is a central modality in modern radiology contributing to diagnostic medicine in almost every medical subspecialty, but particularly in emergency services. To solve the inverse problem of reconstructing anatomical slice images from the raw output the scanner measures, several methods have been developed, with filtered back projection (FBP) and iterative reconstruction (IR) subsequently providing criterion standards. Currently there are new approaches to reconstruction in the field of artificial intelligence utilizing the upcoming possibilities of machine learning (ML), or more specifically, deep learning (DL). METHOD: This review covers the principles of present CT image reconstruction as well as the basic concepts of DL and its implementation in reconstruction. Subsequently commercially available algorithms and current limitations are being discussed. RESULTS AND CONCLUSION: DL is an ML method that utilizes a trained artificial neural network to solve specific problems. Currently two vendors are providing DL image reconstruction algorithms for the clinical routine. For these algorithms, a decrease in image noise and an increase in overall image quality that could potentially facilitate the diagnostic confidence in lesion conspicuity or may translate to dose reduction for given clinical tasks have been shown. One study showed equal diagnostic accuracy in the detection of coronary artery stenosis for DL reconstructed images compared to IR at higher image quality levels. Consequently, a lot more research is necessary and should aim at diagnostic superiority in the clinical context covering a broadness of pathologies to demonstrate the reliability of such DL approaches. KEY POINTS: · Following iterative reconstruction, there is a new approach to CT image reconstruction in the clinical routine using deep learning (DL) as a method of artificial intelligence.. · DL image reconstruction algorithms decrease image noise, improve image quality, and have potential to reduce radiation dose.. · Diagnostic superiority in the clinical context should be demonstrated in future trials.. CITATION FORMAT: · Arndt C, Güttler F, Heinrich A et al. Deep Learning CT Image Reconstruction in Clinical Practice. Fortschr Röntgenstr 2021; 193: 252 - 261.

Heparanase polymorphisms: influence on incidence of hepatic sinusoidal obstruction syndrome in children undergoing allogeneic hematopoietic stem cell transplantation.

PURPOSE: Sinusoidal obstruction syndrome (SOS) is a life-threatening early complication after hematopoietic stem cell transplantation (HSCT), and until now, examinations about the influence of genetic risk factors are extremely rare. The purpose of this study was to identify an association between heparanase (HPSE) single nucleotide polymorphisms (SNPs) and SOS in children undergoing allogeneic HSCT. METHODS: We retrospectively analyzed the distribution of the both HPSE SNPs rs4693608 and rs4364254 and the occurrence of SOS after allogeneic HSCT in 160 children with malignant and non-malignant diseases. RESULTS: Patients with HPSE genotypes GG or AG of rs4693608 (G>A) had a significantly reduced incidence of SOS on day 100 after HSCT compared to patients with genotype AA (4.7 vs. 14.3 %, P = 0.038). In addition, incidence of SOS in patients with genotype CC or CT of rs4364254 (C>T) was significantly decreased in comparison with patients with genotype TT (2.3 vs. 14.7 %, P = 0.004). Interestingly, no patient with genotype CC developed SOS. Because both SNPs co-occur in vivo, we generated subsets: AA-TT, GG-CC, and a group with remaining SNP combinations. We found significant differences between all three patient groups (P = 0.035). Patients with AA-TT showed the highest incidence of SOS (16.7 %), while SOS did not appear in patients with GG-CC (0 %) and residual combinations were numerically in-between (4.9 %). An impact caused by main patient and donor characteristics, established risk factors for SOS, and conditioning regimen could be excluded in multivariate analyses. CONCLUSIONS: HPSE polymorphisms turned out to be significant independent risk factors (P = 0.030) for development of SOS and should be evaluated in further trials.

Prognostic impact of WT1 expression prior to hematopoietic stem cell transplantation in children with malignant hematological diseases.

PURPOSE: Malignant hematological diseases represent the most common pediatric cancer. As they cannot always be cured by chemotherapy alone, leukemia and myelodysplastic syndrome (MDS) are frequent medical indications for hematopoietic stem cell transplantation, yet even this treatment is not capable of preventing relapse for certain. Therefore, molecular markers are used to monitor minimal residual disease (MRD) to be enabled to react early to an impeding relapse. As specific markers are not always available, Wilms' tumor gene 1 (WT1) has been suggested as a universal marker, but has not yet been established clinically. METHODS: We determined the level of WT1 gene expression in 130 children, adolescents and young adults with malignant hematological diseases prior to transplantation and evaluated its impact on patients' outcome. A real-time quantitative RT-PCR was used for this purpose. RESULTS: The relationship between a high level of WT1 and the cumulative incidence of relapse, event-free survival and overall survival proved to be highly significant in univariate and multivariate analyses. Forty-eight percent of all patients with high WT1 levels suffered from a relapse, whereas only eight percent showing normal WT1 levels before transplantation relapsed. The most convincing result was found for acute myeloid leukemia (AML) and MDS. CONCLUSION: We conclude that WT1 expression prior to transplantation qualifies as an independent prognostic factor and should be further evaluated for MRD monitoring. It might especially be useful for patients with AML or MDS missing specific markers.

Wilms tumor gene single nucleotide polymorphism rs16754 predicts a favorable outcome in children with acute lymphoblastic leukemia.

PURPOSE: Wilms tumor gene single nucleotide polymorphism (WT1 SNP) rs16754 has been described as a favorable risk marker in patients with acute myeloid leukemia. Subsequent studies revealed inconsistent results in both adult and pediatric patients. We analyzed its impact on clinical outcome in children with acute lymphoblastic leukemia (ALL) for the first time. METHODS: WT1 SNP rs16754 of 158 children with ALL treated according to ALL Berlin-Frankfurt-Münster treatment trials from 1990 to 2009 and 43 hematopoietic stem cell donors was analyzed by allelic discrimination. WT1 SNP status was correlated with disease characteristics and clinical outcome comparing SNP (WT1(GG/AG)) and wildtype (WT1(AA)). RESULTS: At least one minor allele was found in 23.4 % of patients and 34.9 % of donors (P = 0.07). Distribution of patient and disease characteristics was similar between WT1(GG/AG) and WT1(AA). In multivariate analyses, WT1 SNP was an independent good prognostic marker for cumulative incidence of relapse (CIR WT1(AA) vs. WT1(GG/AG) HR = 3.384, P = 0.021) and event-free survival (EFS; event risk WT1(AA) vs. WT1(GG/AG) HR = 2.503, P = 0.036). Univariate subanalyses of patients who underwent an allogeneic hematopoietic stem cell transplantation revealed more significant differences in CIR (P = 0.017), EFS (P = 0.012), and overall survival (OS; P = 0.017). Donor's WT1 SNP status did not affect outcome. We found no correlation between WT1 SNP and WT1 expression level at diagnosis (P = 0.634). CONCLUSION: WT1 SNP rs16754 predicts improved CIR and EFS. Outcome differences were more prominent in transplanted children. Our findings identify WT1 SNP rs16754 as a favorable risk marker in pediatric ALL which is independent from known risk factors.