Effective training of nanopore callers for epigenetic marks with limited labelled data.

Nanopore sequencing platforms combined with supervised machine learning (ML) have been effective at detecting base modifications in DNA such as 5-methylcytosine (5mC) and N6-methyladenine (6mA). These ML-based nanopore callers have typically been trained on data that span all modifications on all possible DNA [Formula: see text]-mer backgrounds-a complete training dataset. However, as nanopore technology is pushed to more and more epigenetic modifications, such complete training data will not be feasible to obtain. Nanopore calling has historically been performed with hidden Markov models (HMMs) that cannot make successful calls for [Formula: see text]-mer contexts not seen during training because of their independent emission distributions. However, deep neural networks (DNNs), which share parameters across contexts, are increasingly being used as callers, often outperforming their HMM cousins. It stands to reason that a DNN approach should be able to better generalize to unseen [Formula: see text]-mer contexts. Indeed, herein we demonstrate that a common DNN approach (DeepSignal) outperforms a common HMM approach (Nanopolish) in the incomplete data setting. Furthermore, we propose a novel hybrid HMM-DNN approach, amortized-HMM, that outperforms both the pure HMM and DNN approaches on 5mC calling when the training data are incomplete. This type of approach is expected to be useful for calling other base modifications such as 5-hydroxymethylcytosine and for the simultaneous calling of different modifications, settings in which complete training data are not likely to be available.

Graph learning for particle accelerator operations.

Particle accelerators play a crucial role in scientific research, enabling the study of fundamental physics and materials science, as well as having important medical applications. This study proposes a novel graph learning approach to classify operational beamline configurations as good or bad. By considering the relationships among beamline elements, we transform data from components into a heterogeneous graph. We propose to learn from historical, unlabeled data via our self-supervised training strategy along with fine-tuning on a smaller, labeled dataset. Additionally, we extract a low-dimensional representation from each configuration that can be visualized in two dimensions. Leveraging our ability for classification, we map out regions of the low-dimensional latent space characterized by good and bad configurations, which in turn can provide valuable feedback to operators. This research demonstrates a paradigm shift in how complex, many-dimensional data from beamlines can be analyzed and leveraged for accelerator operations.

Supervised training models with or without manual lesion delineation outperform clinicians in distinguishing pulmonary cryptococcosis from lung adenocarcinoma on chest CT.

BACKGROUND: The role of artificial intelligence (AI) in the discrimination between pulmonary cryptococcosis (PC) and lung adenocarcinoma (LA) warrants further research. OBJECTIVES: To compare the performances of AI models with clinicians in distinguishing PC from LA on chest CT. METHODS: Patients diagnosed with confirmed PC or LA were retrospectively recruited from three tertiary hospitals in Guangzhou. A deep learning framework was employed to develop two models: an undelineated supervised training (UST) model utilising original CT images, and a delineated supervised training (DST) model utilising CT images with manual lesion annotations provided by physicians. A subset of 20 cases was randomly selected from the entire dataset and reviewed by clinicians through a network questionnaire. The sensitivity, specificity and accuracy of the models and the clinicians were calculated. RESULTS: A total of 395 PC cases and 249 LA cases were included in the final analysis. The internal validation results for the UST model showed a sensitivity of 85.3%, specificity of 81.0%, accuracy of 83.6% and an area under the curve (AUC) of 0.93. Similarly, the DST model exhibited a sensitivity of 88.2%, specificity of 88.1%, accuracy of 88.2% and an AUC of 0.94. The external validation of the two models yielded AUC values of 0.74 and 0.77, respectively. The average sensitivity, specificity and accuracy of 102 clinicians were determined to be 63.1%, 53.7% and 59.3%, respectively. CONCLUSIONS: Both models outperformed the clinicians in distinguishing between PC and LA on chest CT, with the UST model exhibiting comparable performance to the DST model.

Weakly Supervised MRI Slice-Level Deep Learning Classification of Prostate Cancer Approximates Full Voxel- and Slice-Level Annotation: Effect of Increasing Training Set Size.

BACKGROUND: Weakly supervised learning promises reduced annotation effort while maintaining performance. PURPOSE: To compare weakly supervised training with full slice-wise annotated training of a deep convolutional classification network (CNN) for prostate cancer (PC). STUDY TYPE: Retrospective. SUBJECTS: One thousand four hundred eighty-nine consecutive institutional prostate MRI examinations from men with suspicion for PC (65 ± 8 years) between January 2015 and November 2020 were split into training (N = 794, enriched with 204 PROSTATEx examinations) and test set (N = 695). FIELD STRENGTH/SEQUENCE: 1.5 and 3T, T2-weighted turbo-spin-echo and diffusion-weighted echo-planar imaging. ASSESSMENT: Histopathological ground truth was provided by targeted and extended systematic biopsy. Reference training was performed using slice-level annotation (SLA) and compared to iterative training utilizing patient-level annotations (PLAs) with supervised feedback of CNN estimates into the next training iteration at three incremental training set sizes (N = 200, 500, 998). Model performance was assessed by comparing specificity at fixed sensitivity of 0.97 [254/262] emulating PI-RADS ≥ 3, and 0.88-0.90 [231-236/262] emulating PI-RADS ≥ 4 decisions. STATISTICAL TESTS: Receiver operating characteristic (ROC) and area under the curve (AUC) was compared using DeLong and Obuchowski test. Sensitivity and specificity were compared using McNemar test. Statistical significance threshold was P = 0.05. RESULTS: Test set (N = 695) ROC-AUC performance of SLA (trained with 200/500/998 exams) was 0.75/0.80/0.83, respectively. PLA achieved lower ROC-AUC of 0.64/0.72/0.78. Both increased performance significantly with increasing training set size. ROC-AUC for SLA at 500 exams was comparable to PLA at 998 exams (P = 0.28). ROC-AUC was significantly different between SLA and PLA at same training set sizes, however the ROC-AUC difference decreased significantly from 200 to 998 training exams. Emulating PI-RADS ≥ 3 decisions, difference between PLA specificity of 0.12 [51/433] and SLA specificity of 0.13 [55/433] became undetectable (P = 1.0) at 998 exams. Emulating PI-RADS ≥ 4 decisions, at 998 exams, SLA specificity of 0.51 [221/433] remained higher than PLA specificity at 0.39 [170/433]. However, PLA specificity at 998 exams became comparable to SLA specificity of 0.37 [159/433] at 200 exams (P = 0.70). DATA CONCLUSION: Weakly supervised training of a classification CNN using patient-level-only annotation had lower performance compared to training with slice-wise annotations, but improved significantly faster with additional training data. EVIDENCE LEVEL: 3 TECHNICAL EFFICACY: Stage 2.

Masked image modeling-based boundary reconstruction for 3D medical image segmentation.

Accurate segmentation of 3D medical images is vital for computer-aided diagnosis. However, the complexity of target morphological variations and a scarcity of labeled data make segmentation more challenging. Furthermore, existing models make it difficult to fully and efficiently integrate global and local information, which hinders structured knowledge acquisition. To overcome these challenges, we introduce the TNT Masking Network (TNT-MNet), a groundbreaking transformer-based 3D model that utilizes a transformer-in-transformer (TNT) encoder. For the first time, we present masked image modeling (MIM) in supervised learning, utilizing target boundary regions as masked prediction targets to enhance structured knowledge acquisition. We execute multiscale random masking on inner and outer tokens in online branch to tackle the challenge of segmenting organs and lesion regions with varying structures at multiple scales and to enhance modeling capabilities. In contrast, the target branch utilizes all tokens to guide the online branch to reconstruct the masked tokens. Our experiments suggest that TNT-MNet's performance is comparable, or even better, than state-of-the-art models in three medical image datasets (BTCV, LiTS2017, and BraTS2020) and effectively reduces the dependence on labeled data. The code and models are publicly available at https://github.com/changliu-work/TNT_MNet.

Conserve the open media ecosystem! Legal and ethical considerations when using online repositories for AI training in ecological research.

Ecological researchers who train Artificial Intelligence models using digital media have to be cognizant of legal and ethical implications when sourcing such content from online repositories. The way forward? Complying with Creative Commons licensing requirements, obtaining consent from media creators and adhering to FAIR data principles. Collective action from researchers, repositories, licence providers, and legislators is needed to conserve this complex open media ecosystem. This way, we can continue to develop innovative applications to address pressing ecological issues while maintaining the trust of content creators and respecting the legal and ethical framework of online media use.

Self-supervised deep clustering of single-cell RNA-seq data to hierarchically detect rare cell populations.

Single-cell RNA sequencing (scRNA-seq) is a widely used technique for characterizing individual cells and studying gene expression at the single-cell level. Clustering plays a vital role in grouping similar cells together for various downstream analyses. However, the high sparsity and dimensionality of large scRNA-seq data pose challenges to clustering performance. Although several deep learning-based clustering algorithms have been proposed, most existing clustering methods have limitations in capturing the precise distribution types of the data or fully utilizing the relationships between cells, leaving a considerable scope for improving the clustering performance, particularly in detecting rare cell populations from large scRNA-seq data. We introduce DeepScena, a novel single-cell hierarchical clustering tool that fully incorporates nonlinear dimension reduction, negative binomial-based convolutional autoencoder for data fitting, and a self-supervision model for cell similarity enhancement. In comprehensive evaluation using multiple large-scale scRNA-seq datasets, DeepScena consistently outperformed seven popular clustering tools in terms of accuracy. Notably, DeepScena exhibits high proficiency in identifying rare cell populations within large datasets that contain large numbers of clusters. When applied to scRNA-seq data of multiple myeloma cells, DeepScena successfully identified not only previously labeled large cell types but also subpopulations in CD14 monocytes, T cells and natural killer cells, respectively.

Regularization for Unsupervised Learning of Optical Flow.

Regularization is an important technique for training deep neural networks. In this paper, we propose a novel shared-weight teacher-student strategy and a content-aware regularization (CAR) module. Based on a tiny, learnable, content-aware mask, CAR is randomly applied to some channels in the convolutional layers during training to be able to guide predictions in a shared-weight teacher-student strategy. CAR prevents motion estimation methods in unsupervised learning from co-adaptation. Extensive experiments on optical flow and scene flow estimation show that our method significantly improves on the performance of the original networks and surpasses other popular regularization methods. The method also surpasses all variants with similar architectures and the supervised PWC-Net on MPI-Sintel and on KITTI. Our method shows strong cross-dataset generalization, i.e., our method solely trained on MPI-Sintel outperforms a similarly trained supervised PWC-Net by 27.9% and 32.9% on KITTI, respectively. Our method uses fewer parameters and less computation, and has faster inference times than the original PWC-Net.

SHIP: a computational framework for simulating and validating novel technologies in hardware spiking neural networks.

Investigations in the field of spiking neural networks (SNNs) encompass diverse, yet overlapping, scientific disciplines. Examples range from purely neuroscientific investigations, researches on computational aspects of neuroscience, or applicative-oriented studies aiming to improve SNNs performance or to develop artificial hardware counterparts. However, the simulation of SNNs is a complex task that can not be adequately addressed with a single platform applicable to all scenarios. The optimization of a simulation environment to meet specific metrics often entails compromises in other aspects. This computational challenge has led to an apparent dichotomy of approaches, with model-driven algorithms dedicated to the detailed simulation of biological networks, and data-driven algorithms designed for efficient processing of large input datasets. Nevertheless, material scientists, device physicists, and neuromorphic engineers who develop new technologies for spiking neuromorphic hardware solutions would find benefit in a simulation environment that borrows aspects from both approaches, thus facilitating modeling, analysis, and training of prospective SNN systems. This manuscript explores the numerical challenges deriving from the simulation of spiking neural networks, and introduces SHIP, Spiking (neural network) Hardware In PyTorch, a numerical tool that supports the investigation and/or validation of materials, devices, small circuit blocks within SNN architectures. SHIP facilitates the algorithmic definition of the models for the components of a network, the monitoring of states and output of the modeled systems, and the training of the synaptic weights of the network, by way of user-defined unsupervised learning rules or supervised training techniques derived from conventional machine learning. SHIP offers a valuable tool for researchers and developers in the field of hardware-based spiking neural networks, enabling efficient simulation and validation of novel technologies.

Advancing Brain Metastases Detection in T1-Weighted Contrast-Enhanced 3D MRI Using Noisy Student-Based Training.

The detection of brain metastases (BM) in their early stages could have a positive impact on the outcome of cancer patients. The authors previously developed a framework for detecting small BM (with diameters of <15 mm) in T1-weighted contrast-enhanced 3D magnetic resonance images (T1c). This study aimed to advance the framework with a noisy-student-based self-training strategy to use a large corpus of unlabeled T1c data. Accordingly, a sensitivity-based noisy-student learning approach was formulated to provide high BM detection sensitivity with a reduced count of false positives. This paper (1) proposes student/teacher convolutional neural network architectures, (2) presents data and model noising mechanisms, and (3) introduces a novel pseudo-labeling strategy factoring in the sensitivity constraint. The evaluation was performed using 217 labeled and 1247 unlabeled exams via two-fold cross-validation. The framework utilizing only the labeled exams produced 9.23 false positives for 90% BM detection sensitivity, whereas the one using the introduced learning strategy led to ~9% reduction in false detections (i.e., 8.44). Significant reductions in false positives (>10%) were also observed in reduced labeled data scenarios (using 50% and 75% of labeled data). The results suggest that the introduced strategy could be utilized in existing medical detection applications with access to unlabeled datasets to elevate their performances.

Endoscopy image enhancement method by generalized imaging defect models based adversarial training.

Objective.Smoke, uneven lighting, and color deviation are common issues in endoscopic surgery, which have increased the risk of surgery and even lead to failure.Approach.In this study, we present a new physics model driven semi-supervised learning framework for high-quality pixel-wise endoscopic image enhancement, which is generalizable for smoke removal, light adjustment, and color correction. To improve the authenticity of the generated images, and thereby improve the network performance, we integrated specific physical imaging defect models with the CycleGAN framework. No ground-truth data in pairs are required. In addition, we propose a transfer learning framework to address the data scarcity in several endoscope enhancement tasks and improve the network performance.Main results.Qualitative and quantitative studies reveal that the proposed network outperforms the state-of-the-art image enhancement methods. In particular, the proposed method performs much better than the original CycleGAN, for example, the structural similarity improved from 0.7925 to 0.8648, feature similarity for color images from 0.8917 to 0.9283, and quaternion structural similarity from 0.8097 to 0.8800 in the smoke removal task. Experimental results of the proposed transfer learning method also reveal its superior performance when trained with small datasets of target tasks.Significance.Experimental results on endoscopic images prove the effectiveness of the proposed network in smoke removal, light adjustment, and color correction, showing excellent clinical usefulness.

Supervised training of laparoscopic colorectal cancer resections does not adversely affect short- and long-term outcomes: a Propensity-score-matched cohort study.

BACKGROUND: Supervised training of laparoscopic colorectal cancer surgery to fellows and consultants (trainees) may raise doubts regarding safety and oncological adequacy. This study investigated these concerns by comparing the short- and long-term outcomes of matched supervised training cases to cases performed by the trainer himself. METHODS: A prospective database was analysed retrospectively. All elective laparoscopic colorectal cancer resections in curative intent of adult patients (≥ 18 years) which were performed (non-training cases) or supervised to trainees (training cases) by a single laparoscopic expert surgeon (trainer) were identified. All trainees were specialist surgeons in training for laparoscopic colorectal surgery. Supervised training was standardised. Training cases were 1:1 propensity-score matched to non-training cases using age, American Society of Anesthesiologists (ASA) grade, tumour site (rectum, left and right colon) and American Joint Committee on Cancer (AJCC) tumour stage. The resulting groups were analysed for both short- (operative, oncological, complications) and long-term (time to recurrence, overall and disease-free survival) outcomes. RESULTS: From 10/2006 to 2/2016, a total of 675 resections met the inclusion criteria, of which 95 were training cases. These resections were matched to 95 non-training cases. None of the matched covariates exhibited an imbalance greater than 0.25 (│d│>0.25). There were no significant differences in short- (length of procedure, conversion rate, blood loss, postoperative complications, R0 resections, lymph node harvest) and long-term outcomes. When comparing training cases to non-training cases, 5-year overall and disease-free survival rates were 71.6% (62.4-82.2) versus 81.9% (74.2-90.4) and 70.0% (60.8-80.6) versus 73.6% (64.9-83.3), respectively (not significant). The corresponding hazard ratios (95% confidence intervals, p) were 0.57 (0.32-1.02, p = 0.057) and 0.87 (0.51-1.48, p = 0.61), respectively (univariate Cox proportional hazard model). CONCLUSIONS: Standardised supervised training of laparoscopic colorectal cancer procedures to specialist surgeons may not adversely impact short- and long-term outcomes. This result may also apply to newer surgical techniques as long as standardised teaching methods are followed.

Morphology-aware multi-source fusion-based intracranial aneurysms rupture prediction.

OBJECTIVES: We proposed a new approach to train deep learning model for aneurysm rupture prediction which only uses a limited amount of labeled data. METHOD: Using segmented aneurysm mask as input, a backbone model was pretrained using a self-supervised method to learn deep embeddings of aneurysm morphology from 947 unlabeled cases of angiographic images. Subsequently, the backbone model was finetuned using 120 labeled cases with known rupture status. Clinical information was integrated with deep embeddings to further improve prediction performance. The proposed model was compared with radiomics and conventional morphology models in prediction performance. An assistive diagnosis system was also developed based on the model and was tested with five neurosurgeons. RESULT: Our method achieved an area under the receiver operating characteristic curve (AUC) of 0.823, outperforming deep learning model trained from scratch (0.787). By integrating with clinical information, the proposed model's performance was further improved to AUC = 0.853, making the results significantly better than model based on radiomics (AUC = 0.805, p = 0.007) or model based on conventional morphology parameters (AUC = 0.766, p = 0.001). Our model also achieved the highest sensitivity, PPV, NPV, and accuracy among the others. Neurosurgeons' prediction performance was improved from AUC=0.877 to 0.945 (p = 0.037) with the assistive diagnosis system. CONCLUSION: Our proposed method could develop competitive deep learning model for rupture prediction using only a limited amount of data. The assistive diagnosis system could be useful for neurosurgeons to predict rupture. KEY POINTS: • A self-supervised learning method was proposed to mitigate the data-hungry issue of deep learning, enabling training deep neural network with a limited amount of data. • Using the proposed method, deep embeddings were extracted to represent intracranial aneurysm morphology. Prediction model based on deep embeddings was significantly better than conventional morphology model and radiomics model. • An assistive diagnosis system was developed using deep embeddings for case-based reasoning, which was shown to significantly improve neurosurgeons' performance to predict rupture.

Two-year follow-up after a six-week high-intensity training intervention study with breast cancer patients: physiological, psychological and immunological differences.

PURPOSE: Previously we demonstrated the feasibility of a six-week-long combination of high-intensity interval endurance and strength training (HIT/HIRT) for women with nonmetastatic breast cancer leading to improvements in psychological well-being and performance. Now we report results of a 24-month follow-up. METHODS: Previous intervention (IG, n = 10; 58.7 ± 8.4yrs) and control group (CG, n = 9; 58.8 ± 6.6yrs) were asked for follow-up examinations 12 (T12) and 24 months (T24) after cessation of the supervised training (POST). Medical history, mental well-being, performance and immunological variables were analyzed with respect to intervention start (PRE). RESULTS: IG maximum oxygen consumption (⩒O2peak) 12%-improved POST (p = 0.05) and declined to baseline values T24, while CG ⩒O2peak increased 12% T24 (p = 0.01). IG strength (1RM) increased 31% POST (p < 0.001) and remained above baseline level T24 (p = 0.003), whereas CG 1RM slightly improved T24 (+19%, p = 0.034). IG Anxiety and Depression decreased POST and did not change until T24. IG C-reactive protein decreased POST and increased to pre-exercise levels T24. CG immunological/inflammatory/life quality markers did not change. CONCLUSIONS: Six weeks of HIT/HIRT by breast cancer patients can induce similar beneficial effects like two years of convalescence, but outcomes were unstable and showed a fast backslide in aerobic capacity, activity level and in pro-inflammatory state within 12 months.IMPLICATIONS FOR REHABILITATIONHigh-intensity interval endurance and strength training (HIT/HIRT) for female breast cancer patients was shown to improve psychological well-being and performance, but long-term effects/adherence are unknown.Significant backslides in aerobic capacity, activity level as well as in the pro-inflammatory response after one and two years are observed and should be monitored.Continuous supervision and/or support of breast cancer patients before, during, and after medical care due to poor training adherence when voluntarily executed is recommended.

Comparison of Supervised and Unsupervised Approaches for the Generation of Synthetic CT from Cone-Beam CT.

Due to major artifacts and uncalibrated Hounsfield units (HU), cone-beam computed tomography (CBCT) cannot be used readily for diagnostics and therapy planning purposes. This study addresses image-to-image translation by convolutional neural networks (CNNs) to convert CBCT to CT-like scans, comparing supervised to unsupervised training techniques, exploiting a pelvic CT/CBCT publicly available dataset. Interestingly, quantitative results were in favor of supervised against unsupervised approach showing improvements in the HU accuracy (62% vs. 50%), structural similarity index (2.5% vs. 1.1%) and peak signal-to-noise ratio (15% vs. 8%). Qualitative results conversely showcased higher anatomical artifacts in the synthetic CBCT generated by the supervised techniques. This was motivated by the higher sensitivity of the supervised training technique to the pixel-wise correspondence contained in the loss function. The unsupervised technique does not require correspondence and mitigates this drawback as it combines adversarial, cycle consistency, and identity loss functions. Overall, two main impacts qualify the paper: (a) the feasibility of CNN to generate accurate synthetic CT from CBCT images, which is fast and easy to use compared to traditional techniques applied in clinics; (b) the proposal of guidelines to drive the selection of the better training technique, which can be shifted to more general image-to-image translation.

Development of a supervised exercise programme for immigrant women: Feasibility study of a practical intervention in a physiotherapy setting.

BACKGROUND: While Denmark is facing growing inequality between Danish women and immigrant women in relation to exercise and health, research on interventions and targeted exercise programmes is limited. This study aimed to test the feasibility of a physiotherapeutic supervised exercise programme for immigrant women. METHODS: Inspired by improvement research a programme was developed in cooperation with the immigrant women. The intervention was modified continuously according to the women's wishes and needs. INTERVENTION: Baseline focus-group interviews, completion of questionnaire and physical-strength tests, was followed by a 12-week supervised training period. After completion of the training the participants were re-interviewed and re-tested. RESULTS: Twenty-nine women were recruited to the training programme, and 10 attended follow-up. Mean body mass index was 34 kg/m2. Attendance rate among follow-up tested participants was 70%. The women gained knowledge about their bodies, a healthier lifestyle, and awareness of the importance of active living. CONCLUSION: It was possible to recruit and maintain immigrant women in the exercise programme. This study demonstrated the importance of involving the women in the process, and revealed important factors, such as privacy, a local setting and trust in the physiotherapists.

Prolonged unsupervised Nordic walking and walking exercise following six months of supervision in adults with overweight and obesity: A randomised clinical trial.

BACKGROUND AND AIMS: This study is a randomized trial that examined the effects of 6 months of unsupervised Nordic walking (NW) and walking (W) exercise following 6 months of supervised training in overweight/obese adults. METHODS AND RESULTS: After a 6-month program of diet and supervised training participants (n = 27) of NW (66 ± 7 yrs, body mass index (BMI) 34 ± 5) and W (66 ± 8 yrs, BMI 32 ± 5) group continue the training without supervision for other 6 months. Steps count and mean heart rate (HRmean) were performed in each session; anthropometric and body composition, aerobic capacity and strength of the upper and lower limbs were evaluated at baseline, after 6 months of supervised and 6 months of unsupervised training. In the unsupervised training, monthly sessions and steps count decreased over time in both groups (p < 0.05), with no significant changes in HRmean. Compared to the supervised phase, adherence decreased significantly only in the W group in the last 3 months of unsupervised training. Compared to baseline in both groups BMI did not change, but W group lost total fat; only the NW group maintained (p < 0.05) the gains in arm curl (33%) and chair stand (31%); both groups improved in six-minute walking test (p < 0.05). CONCLUSION: Despite unsupervised training was not effective for a further increase in performance, participants, especially in NW, maintained some of the improvements achieved during the supervision. However, the presence of instructor that guides training, may enhance adherence and health benefits of NW and W exercise. CLINICAL TRIAL REGISTRATION: clinicaltrials.gov Identifier: NCT03212391 (July 11, 2017).

Association of changes in cardiorespiratory fitness with health-related quality of life in young adults with mobility disability: secondary analysis of a randomized controlled trial of mobile app versus supervised training.

BACKGROUND: Young adults with mobility disability report lower health-related quality of life (HRQoL) than their able-bodied peers. This study aims to examine potential differences between the effects of mobile app versus supervised training and the association of cardiorespiratory fitness change with HRQoL in young adults with mobility disability. METHODS: This is a secondary analysis of a parallel randomized controlled trial of a mobile app (n = 55) and a supervised health program (n = 55) that was provided for 12 weeks to 110 adults (18-45 years) with self-perceived mobility disability. Recruitment took place at rehabilitation centers in Stockholm, Sweden. Cardiorespiratory fitness was estimated from the results of a submaximal cycle ergometer test and HRQoL was assessed with the SF-36 questionnaire. Follow up was at 6 weeks, 12 weeks, and 1-year and all examinations were performed by blinded investigators. Between group differences of changes in HRQoL at follow up were estimated in intention-to-treat analysis using linear regression models. Crude and adjusted mixed-effects models estimated the associations between cardiorespiratory fitness change and HRQoL. Stratified analysis by intervention group was also performed. RESULTS: In total, 40/55 from the mobile app group and 49/55 from the supervised training group were included in the intention to treat analysis. No significant differences were observed between the effects of the two interventions on HRQoL. In both crude and adjusted models, cardiorespiratory fitness change was associated with the general health (adjusted ß = 1.30, 95% CI: 0.48, 2.13) and emotional role functioning (adjusted ß = 1.18, 95% CI: 0.11, 2.25) domains of SF-36. After stratification, the associations with general health (adjusted ß = 1.88, 95% CI: 0.87, 2.90) and emotional role functioning (adjusted ß = 1.37, 95% CI: 0.18, 2.57) were present only in the supervised group. CONCLUSION: This study found positive associations between cardiorespiratory fitness change and HRQoL in young adults with mobility disability who received supervised training. The effects of mobile app versus supervised training on HRQoL remain unclear. TRIAL REGISTRATION: International Standard Randomized Controlled Trial Number (ISRCTN) registry ISRCTN22387524 ; Prospectively registered on February 4th, 2018.

EVALUATION OF A SUPERVISED PHYSICAL ACTIVITY PROGRAMME FOR CANCER SURVIVORS: FROM TREATMENT TO TRIATHLON.

OBJECTIVE: It is recommended that cancer survivors incorporate physical activity into their daily lives after in-hospital rehabilitation. However, there is a lack of training programmes focusing on the specific needs of cancer survivors. TriaGO! - an 8-month intervention study of aerobic endurance training for cancer survivors - was therefore examined. The training programme aims to meet the participants' physical needs and provide socio-emotional support, in the form of an exercise programme that challenges participants to aim to compete in an Olympic- distance triathlon (1,000 m swimming, 45 km cycling, 10 km running) after 8 months' of training. METHODS: The TriaGO! training programme was provided to in-hospital rehabilitated cancer survivors (n = 12). Each patient invited a healthy friend or family member to train with them (a so called buddy (n = 12)). The 8-month programme involves supervised training sessions, combining cycling, swimming and running, which progress in frequency, duration and intensity. Physical health was measured at the start, 4 and 8 months, using objective parameters of aerobic fitness, muscular fitness and body composition. RESULTS: A total of 22 out of 24 participants successfully completed the training programme and the triathlon. Both the cancer survivors and their buddies showed significant improvements in physical health. Cancer survivors showed improvements in aerobic fitness, as increases in VO2max and VO2peak of 5.5 ml.kg-1.min-1 and 0.26 ml.min-1 respectively (p <0.0001). Buddies underwent similar significant increases; 5.39 ml.kg-1.min-1 and 0.18 ml.min-1, respectively. CONCLUSION: The TriaGO! training programme introduces the concept of supervised endurance training for cancer survivors. Through measurement of ob-jective parameters, this study demonstrated that significant physical reconditioning is possible in cancer survivors. A supervised programme would be recommended for all cancer patients after in-hospital treatment, in order to facilitate the transition to incorporation of physical activity into daily life.

Advancing Semantic Interoperability of Image Annotations: Automated Conversion of Non-standard Image Annotations in a Commercial PACS to the Annotation and Image Markup.

Sharing radiologic image annotations among multiple institutions is important in many clinical scenarios; however, interoperability is prevented because different vendors' PACS store annotations in non-standardized formats that lack semantic interoperability. Our goal was to develop software to automate the conversion of image annotations in a commercial PACS to the Annotation and Image Markup (AIM) standardized format and demonstrate the utility of this conversion for automated matching of lesion measurements across time points for cancer lesion tracking. We created a software module in Java to parse the DICOM presentation state (DICOM-PS) objects (that contain the image annotations) for imaging studies exported from a commercial PACS (GE Centricity v3.x). Our software identifies line annotations encoded within the DICOM-PS objects and exports the annotations in the AIM format. A separate Python script processes the AIM annotation files to match line measurements (on lesions) across time points by tracking the 3D coordinates of annotated lesions. To validate the interoperability of our approach, we exported annotations from Centricity PACS into ePAD (http://epad.stanford.edu) (Rubin et al., Transl Oncol 7(1):23-35, 2014), a freely available AIM-compliant workstation, and the lesion measurement annotations were correctly linked by ePAD across sequential imaging studies. As quantitative imaging becomes more prevalent in radiology, interoperability of image annotations gains increasing importance. Our work demonstrates that image annotations in a vendor system lacking standard semantics can be automatically converted to a standardized metadata format such as AIM, enabling interoperability and potentially facilitating large-scale analysis of image annotations and the generation of high-quality labels for deep learning initiatives. This effort could be extended for use with other vendors' PACS.