| 1. |
- Ge, Chenjie, 1991, et al.
(författare)
-
Enlarged Training Dataset by Pairwise GANs for Molecular-Based Brain Tumor Classification
- 2020
-
Ingår i: IEEE Access. - 2169-3536 .- 2169-3536. ; 8:1, s. 22560-22570
-
Tidskriftsartikel (refereegranskat)abstract
- This paper addresses issues of brain tumor subtype classification using Magnetic Resonance Images (MRIs) from different scanner modalities like T1 weighted, T1 weighted with contrast-enhanced, T2 weighted and FLAIR images. Currently most available glioma datasets are relatively moderate in size, and often accompanied with incomplete MRIs in different modalities. To tackle the commonly encountered problems of insufficiently large brain tumor datasets and incomplete modality of image for deep learning, we propose to add augmented brain MR images to enlarge the training dataset by employing a pairwise Generative Adversarial Network (GAN) model. The pairwise GAN is able to generate synthetic MRIs across different modalities. To achieve the patient-level diagnostic result, we propose a post-processing strategy to combine the slice-level glioma subtype classification results by majority voting. A two-stage course-to-fine training strategy is proposed to learn the glioma feature using GAN-augmented MRIs followed by real MRIs. To evaluate the effectiveness of the proposed scheme, experiments have been conducted on a brain tumor dataset for classifying glioma molecular subtypes: isocitrate dehydrogenase 1 (IDH1) mutation and IDH1 wild-type. Our results on the dataset have shown good performance (with test accuracy 88.82%). Comparisons with several state-of-the-art methods are also included.
|
|
| 2. |
- Liu, Yuanhua, 1971, et al.
(författare)
-
Considering the importance of user profiles in interface design
- 2009
-
Ingår i: User Interfaces. ; , s. 23-
-
Bokkapitel (övrigt vetenskapligt/konstnärligt)abstract
- User profile is a popular term widely employed during product design processes by industrial companies. Such a profile is normally intended to represent real users of a product. The ultimate purpose of a user profile is actually to help designers to recognize or learn about the real user by presenting them with a description of a real user’s attributes, for instance; the user’s gender, age, educational level, attitude, technical needs and skill level. The aim of this chapter is to provide information on the current knowledge and research about user profile issues, as well as to emphasize the importance of considering these issues in interface design. In this chapter, we mainly focus on how users’ difference in expertise affects their performance or activity in various interaction contexts. Considering the complex interaction situations in practice, novice and expert users’ interactions with medical user interfaces of different technical complexity will be analyzed as examples: one focuses on novice and expert users’ difference when interacting with simple medical interfaces, and the other focuses on differences when interacting with complex medical interfaces. Four issues will be analyzed and discussed: (1) how novice and expert users differ in terms of performance during the interaction; (2) how novice and expert users differ in the perspective of cognitive mental models during the interaction; (3) how novice and expert users should be defined in practice; and (4) what are the main differences between novice and expert users’ implications for interface design. Besides describing the effect of users’ expertise difference during the interface design process, we will also pinpoint some potential problems for the research on interface design, as well as some future challenges that academic researchers and industrial engineers should face in practice.
|
|
| 3. |
- Ali, Muhaddisa Barat, 1986, et al.
(författare)
-
A novel federated deep learning scheme for glioma and its subtype classification
- 2023
-
Ingår i: Frontiers in Neuroscience. - 1662-4548 .- 1662-453X. ; 17
-
Tidskriftsartikel (refereegranskat)abstract
- Background: Deep learning (DL) has shown promising results in molecular-based classification of glioma subtypes from MR images. DL requires a large number of training data for achieving good generalization performance. Since brain tumor datasets are usually small in size, combination of such datasets from different hospitals are needed. Data privacy issue from hospitals often poses a constraint on such a practice. Federated learning (FL) has gained much attention lately as it trains a central DL model without requiring data sharing from different hospitals. Method: We propose a novel 3D FL scheme for glioma and its molecular subtype classification. In the scheme, a slice-based DL classifier, EtFedDyn, is exploited which is an extension of FedDyn, with the key differences on using focal loss cost function to tackle severe class imbalances in the datasets, and on multi-stream network to exploit MRIs in different modalities. By combining EtFedDyn with domain mapping as the pre-processing and 3D scan-based post-processing, the proposed scheme makes 3D brain scan-based classification on datasets from different dataset owners. To examine whether the FL scheme could replace the central learning (CL) one, we then compare the classification performance between the proposed FL and the corresponding CL schemes. Furthermore, detailed empirical-based analysis were also conducted to exam the effect of using domain mapping, 3D scan-based post-processing, different cost functions and different FL schemes. Results: Experiments were done on two case studies: classification of glioma subtypes (IDH mutation and wild-type on TCGA and US datasets in case A) and glioma grades (high/low grade glioma HGG and LGG on MICCAI dataset in case B). The proposed FL scheme has obtained good performance on the test sets (85.46%, 75.56%) for IDH subtypes and (89.28%, 90.72%) for glioma LGG/HGG all averaged on five runs. Comparing with the corresponding CL scheme, the drop in test accuracy from the proposed FL scheme is small (−1.17%, −0.83%), indicating its good potential to replace the CL scheme. Furthermore, the empirically tests have shown that an increased classification test accuracy by applying: domain mapping (0.4%, 1.85%) in case A; focal loss function (1.66%, 3.25%) in case A and (1.19%, 1.85%) in case B; 3D post-processing (2.11%, 2.23%) in case A and (1.81%, 2.39%) in case B and EtFedDyn over FedAvg classifier (1.05%, 1.55%) in case A and (1.23%, 1.81%) in case B with fast convergence, which all contributed to the improvement of overall performance in the proposed FL scheme. Conclusion: The proposed FL scheme is shown to be effective in predicting glioma and its subtypes by using MR images from test sets, with great potential of replacing the conventional CL approaches for training deep networks. This could help hospitals to maintain their data privacy, while using a federated trained classifier with nearly similar performance as that from a centrally trained one. Further detailed experiments have shown that different parts in the proposed 3D FL scheme, such as domain mapping (make datasets more uniform) and post-processing (scan-based classification), are essential.
|
|
| 4. |
- Huvila, J., et al.
(författare)
-
Combined ASRGL1 and p53 immunohistochemistry as an independent predictor of survival in endometrioid endometrial carcinoma
- 2018
-
Ingår i: Gynecologic Oncology. - : Academic Press Inc.. - 0090-8258 .- 1095-6859. ; 149:1, s. 173-180
-
Tidskriftsartikel (refereegranskat)abstract
- Objective: In clinical practise, prognostication of endometrial cancer is based on clinicopathological risk factors. The use of immunohistochemistry-based markers as prognostic tools is generally not recommended and a systematic analysis of their utility as a panel is lacking. We evaluated whether an immunohistochemical marker panel could reliably assess endometrioid endometrial cancer (EEC) outcome independent of clinicopathological information. Methods: A cohort of 306 EEC specimens was profiled using tissue microarray (TMA). Cost- and time-efficient immunohistochemical analysis of well-established tissue biomarkers (ER, PR, HER2, Ki-67, MLH1 and p53) and two new biomarkers (L1CAM and ASRGL1) was carried out. Statistical modelling with embedded variable selection was applied on the staining results to identify minimal prognostic panels with maximal prognostic accuracy without compromising generalizability. Results: A panel including p53 and ASRGL1 immunohistochemistry was identified as the most accurate predictor of relapse-free and disease-specific survival. Within this panel, patients were allocated into high- (5.9%), intermediate- (29.5%) and low- (64.6%) risk groups where high-risk patients had a 30-fold risk (P < 0.001) of dying of EEC compared to the low-risk group. Conclusions: P53 and ASRGL1 immunoprofiling stratifies EEC patients into three risk groups with significantly different outcomes. This simple and easily applicable panel could provide a useful tool in EEC risk stratification and guiding the allocation of treatment modalities.
|
|
| 5. |
- Rosengren, Annika, 1951, et al.
(författare)
-
COVID-19 in people aged 18–64 in Sweden in the first year of the pandemic: Key factors for severe disease and death
- 2022
-
Ingår i: Global Epidemiology. - : Elsevier BV. - 2590-1133. ; 4
-
Tidskriftsartikel (refereegranskat)abstract
- Background: Studies on risk factors for severe COVID-19 in people of working age have generally not included non-working persons or established population attributable fractions (PAFs) for occupational and other factors. Objectives: We describe the effect of job-related, sociodemographic, and other exposures on the incidence, relative risks and PAFs of severe COVID-19 in individuals aged 18–64. Methods: We conducted a registry-based study in Swedish citizens aged 18–64 from 1 January 2020 to 1 February 2021 with respect to COVID-19-related hospitalizations and death. Results: Of 6,205,459 persons, 272,043 (7.5%) were registered as infected, 3399 (0.05%) needed intensive care, and 620 (0.01%) died, with an estimated case fatality rate of 0.06% over the last 4-month period when testing was adequate. Non-Nordic origin was associated with a RR for need of intensive care of 3·13, 95%CI 2·91–3·36, and a PAF of 32·2% after adjustment for age, sex, work, region and comorbidities. In a second model with occupation as main exposure, and adjusted for age, sex, region, comorbidities and origin, essential workers had an RR of 1·51, 95%CI, 1·35–1·6, blue-collar workers 1·18, 95%CI 1·06–1·31, school staff 1·21, 95%CI 1·01–1·46, and health and social care workers 1·89, 95%CI 1·67–2·135) compared with people able to work from home, with altogether about 13% of the PAF associated with these occupations. Essential workers and blue-collar workers, but no other job categories had higher risk of death, adjusted RRs of 1·79, 95%CI 1·34–2·38 and 1·37, 95%CI 1·04–1·81, with adjusted PAFs of altogether 9%. Conclusion: Among people of working age in Sweden, overall mortality and case fatality were low. Occupations that require physical presence at work were associated with elevated risk of needing intensive care for COVID-19, with 14% cases attributable to this factor, and 9% of deaths.
|
|
| 6. |
- Hartvigsson, Olle, 1991, et al.
(författare)
-
Differences between Arterial and Venous Umbilical Cord Plasma Metabolome and Association with Parity
- 2022
-
Ingår i: Metabolites. - : MDPI AG. - 2218-1989 .- 2218-1989. ; 12:2
-
Tidskriftsartikel (refereegranskat)abstract
- Umbilical cord blood is frequently used in health monitoring of the neonate. Results may be affected by the proportion of arterial and venous cord blood, the venous blood coming from the mother to supply oxygen and nutrients to the infant, and the arterial carrying waste products from the fetus. Here, we sampled arterial and venous umbilical cords separately from 48 newly delivered infants and examined plasma metabolomes using GC-MS/MS metabolomics. We investigated differences in metabolomes between arterial and venous blood and their associations with gestational length, birth weight, sex, and whether the baby was the first born or not, as well as maternal age and BMI. Using multilevel random forest analysis, a classification rate of 79% was achieved for arteriovenous differences (p = 0.004). Several monosaccharides had higher concentrations in the arterial cord plasma while amino acids were higher in venous plasma, suggesting that the main differences in the measured arterial and venous plasma metabolomes are related to amino acid and energy metabolism. Venous cord plasma metabolites related to energy metabolism were positively associated with parity (77% classification rate, p = 0.004) while arterial cord plasma metabolites were not. This underlines the importance of defining cord blood type for metabolomic studies.
|
|
| 7. |
- Li, Peishun, 1988, et al.
(författare)
-
Metabolic Alterations in Older Women With Low Bone Mineral Density Supplemented With Lactobacillus reuteri
- 2021
-
Ingår i: JBMR Plus. - : Wiley. - 2473-4039. ; 5:4
-
Tidskriftsartikel (refereegranskat)abstract
- JBMR Plus published by Wiley Periodicals LLC on behalf of American Society for Bone and Mineral Research. Osteoporosis and its associated fractures are highly prevalent in older women. Recent studies have shown that gut microbiota play important roles in regulating bone metabolism. A previous randomized controlled trial (RCT) found that supplementation with Lactobacillus reuteri ATCC PTA 6475 (L.reuteri) led to substantially reduced bone loss in older women with low BMD. However, the total metabolic effects of L. reuteri supplementation on older women are still not clear. In this study, a post hoc analysis (not predefined) of serum metabolomic profiles of older women from the previous RCT was performed to investigate the metabolic dynamics over 1 year and to evaluate the effects of L. reuteri supplementation on human metabolism. Distinct segregation of the L. reuteri and placebo groups in response to the treatment was revealed by partial least squares-discriminant analysis. Although no individual metabolite was differentially and significantly associated with treatment after correction for multiple testing, 97 metabolites responded differentially at any one time point between L. reuteri and placebo groups (variable importance in projection score >1 and p value <0.05). These metabolites were involved in multiple processes, including amino acid, peptide, and lipid metabolism. Butyrylcarnitine was particularly increased at all investigated time points in the L. reuteri group compared with placebo, indicating that the effects of L. reuteri on bone loss are mediated through butyrate signaling. Furthermore, the metabolomic profiles in a case (low BMD) and control population (high BMD) of elderly women were analyzed to confirm the associations between BMD and the identified metabolites regulated by L. reuteri supplementation. The amino acids, especially branched-chain amino acids, showed association with L. reuteri treatment and with low BMD in older women, and may serve as potential therapeutic targets. © 2021 The Authors. JBMR Plus published by Wiley Periodicals LLC on behalf of American Society for Bone and Mineral Research.
|
|
| 8. |
- Religa, D., et al.
(författare)
-
SveDem, the Swedish Dementia Registry - A tool for improving the quality of diagnostics, treatment and care of dementia patients in clinical practice
- 2015
-
Ingår i: PLoS ONE. - : Public Library of Science (PLoS). - 1932-6203. ; 10:2
-
Tidskriftsartikel (refereegranskat)abstract
- Background: The Swedish Dementia Registry (SveDem) was developed with the aim to improve the quality of diagnostic work-up, treatment and care of patients with dementia disorders in Sweden. Methods: SveDem is an internet based quality registry where several indicators can be followed over time. It includes information about the diagnostic work-up, medical treatment and community support (www.svedem.se). The patients are diagnosed and followed-up yearly in specialist units, primary care centres or in nursing homes. Results: The database was initiated in May 2007 and covers almost all of Sweden. There were 28 722 patients registered with a mean age of 79.3 years during 2007-2012. Each participating unit obtains continuous online statistics from its own registrations and they can be compared with regional and national data. A report from SveDem is published yearly to inform medical and care professionals as well as political and administrative decision-makers about the current quality of diagnostics, treatment and care of patients with dementia disorders in Sweden. Conclusion: SveDem provides knowledge about current dementia care in Sweden and serves as a framework for ensuring the quality of diagnostics, treatment and care across the country. It also reflects changes in quality dementia care over time. Data from SveDem can be used to further develop the national guidelines for dementia and to generate new research hypotheses.
|
|
| 9. |
|
|
| 10. |
- Grynne, A., et al.
(författare)
-
Women's experience of the health information process involving a digital information tool before commencing radiation therapy for breast cancer : a deductive interview study
- 2023
-
Ingår i: BMC Health Services Research. - : BioMed Central (BMC). - 1472-6963. ; 23:1
-
Tidskriftsartikel (refereegranskat)abstract
- BACKGROUND: Individuals undergoing radiation therapy for breast cancer frequently request information before, throughout and after the treatment as a means to reduce distress. Nevertheless, the provision of information to meet individuals needs from their level of health literacy is often overlooked. Thus, individuals information needs are often unmet, leading to reports of discontent. Internet and digital information technology has significantly augmented the available information and changed the way in which persons accesses and comprehends information. As health information is no longer explicitly obtained from healthcare professionals, it is essential to examine the sequences of the health information process in general, and in relation to health literacy. This paper reports on qualitative interviews, targeting women diagnosed with breast cancer who were given access to a health information technology tool, Digi-Do, before commencing radiation therapy, during, and after treatment. METHODS: A qualitative research design, inspired by the integrated health literacy model, was chosen to enable critical reflection by the participating women. Semi-structured interviews were conducted with 15 women with access to a digital information tool, named Digi-Do, in addition to receiving standard information (oral and written) before commencing radiation therapy, during, and after treatment. A deductive thematic analysis process was conducted. RESULTS: The results demonstrate how knowledge, competence, and motivation influence women's experience of the health information process. Three main themes were found: Meeting interactive and personal needs by engaging with health information; Critical recognition of sources of information; and Capability to communicate comprehended health information. The findings reflect the women's experience of the four competencies: to access, understand, appraise, and apply, essential elements of the health information process. CONCLUSIONS: We can conclude that there is a need for tailored digital information tools, such as the Digi-Do, to enable iterative access and use of reliable health information before, during and after the radiation therapy process. The Digi-Do can be seen as a valuable complement to the interpersonal communication with health care professionals, facilitating a better understanding, and enabling iterative access and use of reliable health information before, during and after the radiotherapy treatment. This enhances a sense of preparedness before treatment starts.
|
|
