Interaction between clinicians and artificial intelligence to detect fetal atrioventricular septal defects on ultrasound: how can we optimize collaborative performance?

OBJECTIVES: Artificial intelligence (AI) has shown promise in improving the performance of fetal ultrasound screening in detecting congenital heart disease (CHD). The effect of giving AI advice to human operators has not been studied in this context. Giving additional information about AI model workings, such as confidence scores for AI predictions, may be a way of further improving performance. Our aims were to investigate whether AI advice improved overall diagnostic accuracy (using a single CHD lesion as an exemplar), and to determine what, if any, additional information given to clinicians optimized the overall performance of the clinician-AI team. METHODS: An AI model was trained to classify a single fetal CHD lesion (atrioventricular septal defect (AVSD)), using a retrospective cohort of 121 130 cardiac four-chamber images extracted from 173 ultrasound scan videos (98 with normal hearts, 75 with AVSD); a ResNet50 model architecture was used. Temperature scaling of model prediction probability was performed on a validation set, and gradient-weighted class activation maps (grad-CAMs) produced. Ten clinicians (two consultant fetal cardiologists, three trainees in pediatric cardiology and five fetal cardiac sonographers) were recruited from a center of fetal cardiology to participate. Each participant was shown 2000 fetal four-chamber images in a random order (1000 normal and 1000 AVSD). The dataset comprised 500 images, each shown in four conditions: (1) image alone without AI output; (2) image with binary AI classification; (3) image with AI model confidence; and (4) image with grad-CAM image overlays. The clinicians were asked to classify each image as normal or AVSD. RESULTS: A total of 20 000 image classifications were recorded from 10 clinicians. The AI model alone achieved an accuracy of 0.798 (95% CI, 0.760-0.832), a sensitivity of 0.868 (95% CI, 0.834-0.902) and a specificity of 0.728 (95% CI, 0.702-0.754), and the clinicians without AI achieved an accuracy of 0.844 (95% CI, 0.834-0.854), a sensitivity of 0.827 (95% CI, 0.795-0.858) and a specificity of 0.861 (95% CI, 0.828-0.895). Showing a binary (normal or AVSD) AI model output resulted in significant improvement in accuracy to 0.865 (P < 0.001). This effect was seen in both experienced and less-experienced participants. Giving incorrect AI advice resulted in a significant deterioration in overall accuracy, from 0.761 to 0.693 (P < 0.001), which was driven by an increase in both Type-I and Type-II errors by the clinicians. This effect was worsened by showing model confidence (accuracy, 0.649; P < 0.001) or grad-CAM (accuracy, 0.644; P < 0.001). CONCLUSIONS: AI has the potential to improve performance when used in collaboration with clinicians, even if the model performance does not reach expert level. Giving additional information about model workings such as model confidence and class activation map image overlays did not improve overall performance, and actually worsened performance for images for which the AI model was incorrect. © 2024 The Authors. Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.

Automated fetal brain segmentation from 2D MRI slices for motion correction.

Motion correction is a key element for imaging the fetal brain in-utero using Magnetic Resonance Imaging (MRI). Maternal breathing can introduce motion, but a larger effect is frequently due to fetal movement within the womb. Consequently, imaging is frequently performed slice-by-slice using single shot techniques, which are then combined into volumetric images using slice-to-volume reconstruction methods (SVR). For successful SVR, a key preprocessing step is to isolate fetal brain tissues from maternal anatomy before correcting for the motion of the fetal head. This has hitherto been a manual or semi-automatic procedure. We propose an automatic method to localize and segment the brain of the fetus when the image data is acquired as stacks of 2D slices with anatomy misaligned due to fetal motion. We combine this segmentation process with a robust motion correction method, enabling the segmentation to be refined as the reconstruction proceeds. The fetal brain localization process uses Maximally Stable Extremal Regions (MSER), which are classified using a Bag-of-Words model with Scale-Invariant Feature Transform (SIFT) features. The segmentation process is a patch-based propagation of the MSER regions selected during detection, combined with a Conditional Random Field (CRF). The gestational age (GA) is used to incorporate prior knowledge about the size and volume of the fetal brain into the detection and segmentation process. The method was tested in a ten-fold cross-validation experiment on 66 datasets of healthy fetuses whose GA ranged from 22 to 39 weeks. In 85% of the tested cases, our proposed method produced a motion corrected volume of a relevant quality for clinical diagnosis, thus removing the need for manually delineating the contours of the brain before motion correction. Our method automatically generated as a side-product a segmentation of the reconstructed fetal brain with a mean Dice score of 93%, which can be used for further processing.

Towards automated extraction of 2D standard fetal head planes from 3D ultrasound acquisitions: A clinical evaluation and quality assessment comparison.

INTRODUCTION: Clinical evaluation of deep learning (DL) tools is essential to compliment technical accuracy metrics. This study assessed the image quality of standard fetal head planes automatically-extracted from three-dimensional (3D) ultrasound fetal head volumes using a customised DL-algorithm. METHODS: Two observers retrospectively reviewed standard fetal head planes against pre-defined image quality criteria. Forty-eight images (29 transventricular, 19 transcerebellar) were selected from 91 transabdominal fetal scans (mean gestational age = 26 completed weeks, range = 20+5-32+3 weeks). Each had two-dimensional (2D) manually-acquired (2D-MA), 3D operator-selected (3D-OS) and 3D-DL automatically-acquired (3D-DL) images. The proportion of adequate images from each plane and modality, and the number of inadequate images per plane was compared for each method. Inter and intra-observer agreement of overall image quality was calculated. RESULTS: Sixty-seven percent of 3D-OS and 3D-DL transventricular planes were adequate quality. Forty-five percent of 3D-OS and 55% of 3D-DL transcerebellar planes were adequate. Seventy-one percent of 3D-OS and 86% of 3D-DL transventricular planes failed with poor visualisation of intra-cranial structures. Eighty-six percent of 3D-OS and 80% of 3D-DL transcerebellar planes failed due to inadequate visualisation of cerebellar hemispheres. Image quality was significantly different between 2D and 3D, however, no significant difference between 3D-modalities was demonstrated (p < 0.005). Inter-observer agreement of transventricular plane adequacy was moderate for both 3D-modalities, and weak for transcerebellar planes. CONCLUSION: The 3D-DL algorithm can automatically extract standard fetal head planes from 3D-head volumes of comparable quality to operator-selected planes. Image quality in 3D is inferior to corresponding 2D planes, likely due to limitations with 3D-technology and acquisition technique. IMPLICATIONS FOR PRACTICE: Automated image extraction of standard planes from US-volumes could facilitate use of 3DUS in clinical practice, however image quality is dependent on the volume acquisition technique.

Deregulated expression of fat and muscle genes in B-cell chronic lymphocytic leukemia with high lipoprotein lipase expression.

Lipoprotein lipase (LPL) is a prognostic marker in B-cell chronic lymphocytic leukemia (B-CLL) related to immunoglobulin V(H) gene (IgV(H))mutational status. We determined gene expression profiles using Affymetrix U133A GeneChips in two groups of B-CLLs selected for either high ('LPL+', n=10) or low ('LPL-', n=10) LPL mRNA expression. Selected genes were verified by real-time PCR in an extended patient cohort (n=42). A total of 111 genes discriminated LPL+ from LPL- B-CLLs. Of these, the top three genes associated with time to first treatment were Septin10, DMD and Gravin (P

Impact of cytogenetics on the prognosis of adults with de novo AML in first relapse.

Karyotype is an important prognostic factor in patients with newly diagnosed acute myeloblastic leukaemia (AML). The prognostic value of cytogenetics on the outcome of patients with AML in relapse has not yet been well defined. We analysed the clinical outcome of 152 patients with de novo, chemotherapy-treated AML in first relapse according to the cytogenetic classification of the United Kingdom Medical Research Council. The rate of second complete remission (CR) (88, 64 and 36%) and the probability of survival at 3 years (43, 18 and 0%) were significantly different between the favourable, intermediate and adverse cytogenetic risk groups, respectively. Compared to the favourable group, the relative risk (RR) of death (multivariate analyses) was 2.6 (confidence interval (CI): 1.5-4.4, P<0.001) for the intermediate and 3.7 (CI: 1.7-7.9, P=0.001) for the adverse group. The prognostic value of the duration of first CR was confirmed (RR of death: 2.0 (CI: 1.0-4.0) for each additional year in first CR), whereas the FLT3 mutation obtained at diagnosis did not markedly influence the outcome of patients with AML in relapse. In conclusion, our results indicate that both karyotype and the duration of first CR are independent prognostic factors for patients with de novo AML in first relapse.

[Comparison of three outpatient therapy forms for treatment of chronic low back pain-- findings of a multicentre, cluster randomized study]. / Vergleich dreier ambulanter Therapieformen zur Behandlung chronischer Rückenschmerzen-- Ergebnisse einer multizentrischen, clusterrandomisierten Studie.

The AOK Baden-Württemberg health insurance fund initiated a study on the outpatient rehabilitation of patients with chronic low back pain, aimed at improving the treatment concept for its insurees with chronic low back pain (START). This model project was scientifically guided by the Hochrhein-Institute in Bad Säckingen. The paper compares the effectiveness of Enhanced Outpatient Physiotherapy (Erweiterte Ambulante Physiotherapie, EAP), Outpatient Rehabilitation (Ambulante Rehabilitation, AR) and Medical Training Therapy (Medizinische Trainingstherapie, MTT) in patients with low back pain. In seven regions in Baden-Württemberg, one of these three intervention forms was provided to the patients. A total of 1,274 patients were included in the study. The AOK Baden-Württemberg patients receiving treatment in one of the three intervention forms were seriously restricted in both the physical dimension of their health status and in their physical mobility in everyday life and at the workplace. Besides, they frequently reported considerable psychosocial strain. The three interventions led to significant and relevant decreases in pain intensity and to an improved health-related quality of life. There were no significant differences between the various treatments in terms of effectiveness. The patients shared an equally high satisfaction with the treatment received. In MTT, the total therapy length of 15 weeks was by far longer than in AR and EAP (about 5 and 8 weeks). Unlike AR and, in parts, EAP, patients may continue to work while participating in MTT. Therefore an immediate therapy start within a week was more likely possible in MTT (59 %) than in AR (10 %) or EAP (23 %). In evaluating the results a number of restrictions have to be considered. Nevertheless, based on our research findings, the following can be concluded: MTT is a suitable therapy concept in patients with low back pain characterized by a rapid start and-- compared to the other two concepts-- by lower therapy costs. MTT might represent a meaningful therapy element also in new forms of provision such as integrated services. Patients showing severe psychosocial strain should be assigned to an interdisciplinary therapy as it is provided by inpatient and outpatient rehabilitation facilities. For effective differential assignment to the various programmes, realization of a standardized rehab assessment is an important precondition.

Monitoring minimal residual disease in AML: the right time for real time.

Detection of minimal residual disease (MRD) by polymerase chain reaction (PCR) has become an essential tool for molecular monitoring of acute myeloid leukemia (AML). Currently, specific translocation markers are available for 40-50% of AMLs. Expression markers may widen this spectrum to 70-90%. Quantitative PCR (Q-PCR, real-time PCR) is now as sensitive as conventional two-step PCR and could improve as well as facilitate clinical decision-making. Q-PCR has been applied to a variety of molecular markers, delineating threshold levels early after induction therapy, for postinduction monitoring, as well as for the detection of relapse. For most markers, lack of decline of transcript levels by less than 2 logs after chemotherapy has been established as a poor prognostic sign. Moreover, increases in transcript levels are almost invariably associated with relapse. However, the predictive value of PCR negativity after chemotherapy is not as clear. The major tasks for the future will be standardization of Q-PCR techniques, exact definition of threshold levels, and monitoring schedules in bone marrow (BM) and peripheral blood (PB), as well as investigation of novel markers found by microarray analysis.

[Evolution of a quality assurance programme for physiotherapy schools - results of the first quality inspections]. / Entwicklung eines Qualitätssicherungsprogramms für Physiotherapieschulen - Ergebnisse der ersten Qualitätsprüfungen.

In cooperation with the Hochrhein-Institute for Research in Rehabilitation (HRI), the Association for Assuring the Quality of Education in Physiotherapy Schools in Germany (ISQ) has developed a quality assurance programme for physiotherapy schools. It aims at assessing the quality of physiotherapy schools in Germany, and to award a quality seal based on compliance with defined criteria. First, a catalogue of quality features and criteria relevant for education in physiotherapy was developed. It is based on the analysis of questionnaires that had been sent to all German physiotherapy schools, to selected physiotherapists and leading physiotherapists in hospitals, to competent federal authorities, and to three school-classes with group discussions. The persons addressed named 360 different quality features. They were collected in a catalogue, revised in a multi-stage Delphi procedure, and approved consensually. The final criteria were transformed into basic quality requirements, and formulated as a check-list. Assessment of the quality features is carried out by trained visitors. In addition, the satisfaction of students is assessed with a questionnaire. The results of the interviews and the questionnaires are fed back to the schools in a quality report. Schools meeting all basic quality requirements are awarded the seal of quality. The seal is valid for three years. Since January 2003, this procedure is available for all schools in Germany. Until September 2002, a pretest of visitations and student questionnaires had been carried out with 31 member schools of the ISQ; according to the resulting quality reports, none of these schools would instantly be awarded the quality seal. In all, more than half of the schools do not meet 10 of the 42 basic criteria. Fundamental deficiencies have been found in the documentation pertaining to supervision of practical training. In terms of training, further training and professional development of their teachers and associated professors, needs for improvement could be shown in more than 66 % of all visited schools. Only 9 of 31 schools could produce a written syllabus. Additionally, the requirements of teachers conferences and equipment of libraries were not met by the majority. A general problem among the schools is inadequate documentation in many fields.