| 1. |
|
|
| 2. |
- Sarve, Hamid, et al.
(författare)
-
Quantification of bone remodeling in the proximity of implants
- 2007
-
Ingår i: Proceedings of the 12th International Conference on Computer Analysis of Images and Patterns (CAIP07). - Berlin : Springer. - 9783540742715 ; , s. 253-260, s. 253-260
-
Konferensbidrag (refereegranskat)abstract
- In histomorphometrical investigations of bone tissue modeling around screw-shaped implants, the manual measurements of bone area and bone-implant contact length around the implant are time consuming and subjective. In this paper we propose an automatic image analysis method for such measurements. We evaluate different discriminant analysis methods and compare the automatic method with the manual one. The results show that the principal difference between the two methods occurs in length estimation, whereas the area measurement does not differ significantly. A major factor behind the dissimilarities in the results is believed to be misclassification of staining artifacts by the automatic method.
|
|
| 3. |
- Lundberg, Peter, et al.
(författare)
-
Kvantifiering av leversteatos: diagnostisk utvärdering av protonmagnetresonansspektroskopi jämfört med histologiska metoder
- 2016
-
Konferensbidrag (refereegranskat)abstract
- BakgrundLeversteatos är den vanligaste manifestationen av leversjukdom i västvärlden. Leverbiopsi med semikvantitativ histologisk gradering är referensmetod vid gradering av leversteatos. Med protonmagnetsresonansspektroskopi (1H-MRS), en metod som föreslagits ersätta leverbiopsi för värdering av steatos, kan leverns innehåll av triglycerider mätas icke-invasivt. Triglyceridinnehåll >5,00 % används ofta som ett diagnostiskt kriterium för leversteatos vid undersökning med 1H-MRS. Syftet med studien var att jämföra 1H-MRS med semikvantitativ histologisk steatosgradering och kvantitativ histologisk steatosmätning.MetodPatienter remitterade för utredning av förhöjda leverenzymer in-kluderades i studien. Samtliga patienter genomgick klinisk undersökning, laboratorieprovtagning samt 1H-MRS direkt följd av leverbiopsi. För konventionell histologisk semikvantitativ gradering av steatos användes kriterierna utarbetade av Brunt och medarbetare. Kvantitativ mätning av fett i biopsierna utfördes genom att med hjälp av stereologisk punkträkning (SPC) mäta andelen av ytan som innehöll fettvakuoler.ResultatI studien inkluderades 94 patienter, varav 37 hade icke-alkoholor-sakad fettleversjukdom (NAFLD), 49 hade andra leversjukdomar och 8 hade normal leverbiopsi. En stark korrelation noterades mel-lan 1H-MRS och SPC (r=0,92, p<0,0001; к=0.82). Korrelationen mellan 1H-MRS och Brunts kriterier (к=0.26) samt mellan SPC och Brunts kriterier (к=0.38) var betydligt sämre. När patologens gradering (Brunts kriterier) användes som referensmetod för diag-nos av leversteatos så hade alla patienter med triglyceridinnehåll >5,00 % mätt med 1H-MRS steatos (specificitet 100 %). Emellertid hade 22 av 69 patienter med triglyceridinnehåll ≤5,00 % också le-versteatos enligt Brunts kriterier (sensitivitet 53 %). Motsvarande siffror när man använde gränsvärdet 3,02 % var sensitivitet 79 % och specificitet 100 %. Vid ytterligare reduktion av gränsvärdet för triglyceridinnehåll till 2,00 % ökade sensitiviteten till 87 % med upprätthållande av hög specificitet (94 %).Slutsats1H-MRS och SPC uppvisade en mycket hög korrelation vid kvantifiering av leversteatos. SPC borde därför föredras framför Brunts kriterier när noggrann histologisk kvantifiering av leversteatos är önskvärd. Många patienter kan ha histologisk leversteatos trots triglyceridinnehåll ≤5,00 % mätt med 1H-MRS. Gränsvärdet för diagnostisering av leversteatos med 1H-MRS bör därför reduceras.
|
|
| 4. |
- Hellström Ängerud, Karin, et al.
(författare)
-
Differences in symptoms in relation to myocardial infarction.
- 2016
-
Konferensbidrag (övrigt vetenskapligt/konstnärligt)abstract
- Background: In myocardial infarction (MI) rapid diagnosis and treatment is crucial for the prognosis. Previous research has found that symptom presentation influence pre hospital delay times but studies about differences in MI symptoms between patients with ST-elevation myocardial infarction (STEMI) and non ST-elevation myocardial infarction (NSTEMI) are sparse and inconclusive. To enhance the understanding of symptom presentation in regard to MI type, we aimed to describe symptoms in relation to MI type and to find predictors of STEMI versus NSTEMI in patients with MI.Methods: Patients with MI (n=694) from the SymTime study were included. SymTime was a multicentre cross-sectional study of symptoms and actions in the prehospital phase of MI and data were collected using a previously validated questionnaire administered to MI patients within 24 h of admission to hospital.Results: Patients with STEMI were younger, more often men and smokers. Patients with NSTEMI were more likely to have a history of hypertension, MI and stroke. Chest pain was the most common symptom in both groups. Pain, discomfort, or pressure located in the jaw or teeth, vertigo/pre-syncope, cold sweat and nausea/vomiting were significantly more frequent in patients with STEMI (Table 1). In a multivariate logistic regression model patients with STEMI were more likely to present with cold sweat (OR 4.13, 95% CI 2.71–6.29) jaw pain (OR 2.14, 95% CI 1.02–4.50), and nausea (OR 2.01, 95% CI 1.20–3.33), and less likely to have a history of stroke (OR 0.35, 95% CI 0.15–0.84), fluctuating symptoms (OR 0.54, 95% CI 0.36–0.83) and anxiety (OR 0.54, 95% CI 0.32–0.92) compared to patients with NSTEMI.Conclusion: Patients with STEMI differed significantly from those with NSTEMI regarding symptom presentation. This knowledge is important for health care personnel to recognize symptoms alarming for STEMI when evaluating patients with MI symptoms.
|
|
| 5. |
- Ali, Muhaddisa Barat, 1986, et al.
(författare)
-
Multi-stream Convolutional Autoencoder and 2D Generative Adversarial Network for Glioma Classification
- 2019
-
Ingår i: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). - Cham : Springer International Publishing. - 1611-3349 .- 0302-9743. ; 11678 LNCS, s. 234-245
-
Konferensbidrag (refereegranskat)abstract
- Diagnosis and timely treatment play an important role in preventing brain tumor growth. Deep learning methods have gained much attention lately. Obtaining a large amount of annotated medical data remains a challenging issue. Furthermore, high dimensional features of brain images could lead to over-fitting. In this paper, we address the above issues. Firstly, we propose an architecture for Generative Adversarial Networks to generate good quality synthetic 2D MRIs from multi-modality MRIs (T1 contrast-enhanced, T2, FLAIR). Secondly, we propose a deep learning scheme based on 3-streams of Convolutional Autoencoders (CAEs) followed by sensor information fusion. The rational behind using CAEs is that it may improve glioma classification performance (as comparing with conventional CNNs), since CAEs offer noise robustness and also efficient feature reduction hence possibly reduce the over-fitting. A two-round training strategy is also applied by pre-training on GAN augmented synthetic MRIs followed by refined-training on original MRIs. Experiments on BraTS 2017 dataset have demonstrated the effectiveness of the proposed scheme (test accuracy 92.04%). Comparison with several exiting schemes has provided further support to the proposed scheme.
|
|
| 6. |
- Ge, Chenjie, 1991, et al.
(författare)
-
Cross-Modality Augmentation of Brain Mr Images Using a Novel Pairwise Generative Adversarial Network for Enhanced Glioma Classification
- 2019
-
Ingår i: Proceedings - International Conference on Image Processing, ICIP. - 1522-4880.
-
Konferensbidrag (refereegranskat)abstract
- © 2019 IEEE. Brain Magnetic Resonance Images (MRIs) are commonly used for tumor diagnosis. Machine learning for brain tumor characterization often uses MRIs from many modalities (e.g., T1-MRI, Enhanced-T1-MRI, T2-MRI and FLAIR). This paper tackles two issues that may impact brain tumor characterization performance from deep learning: insufficiently large training dataset, and incomplete collection of MRIs from different modalities. We propose a novel pairwise generative adversarial network (GAN) architecture for generating synthetic brain MRIs in missing modalities by using existing MRIs in other modalities. By improving the training dataset, we aim to mitigate the overfitting and improve the deep learning performance. Main contributions of the paper include: (a) propose a pairwise generative adversarial network (GAN) for brain image augmentation via cross-modality image generation; (b) propose a training strategy to enhance the glioma classification performance, where GAN-augmented images are used for pre-training, followed by refined-training using real brain MRIs; (c) demonstrate the proposed method through tests and comparisons of glioma classifiers that are trained from mixing real and GAN synthetic data, as well as from real data only. Experiments were conducted on an open TCGA dataset, containing 167 subjects for classifying IDH genotypes (mutation or wild-type). Test results from two experimental settings have both provided supports to the proposed method, where glioma classification performance has consistently improved by using mixed real and augmented data (test accuracy 81.03%, with 2.57% improvement).
|
|
| 7. |
- Salford, Leif, et al.
(författare)
-
The mammalian brain in the electromagnetic fields designed by man with special reference to blood-brain barrier function, neuronal damage and possible physical mechanisms
- 2008
-
Ingår i: PROGRESS OF THEORETICAL PHYSICS SUPPLEMENT. - 0375-9687. ; :173, s. 283-309
-
Konferensbidrag (refereegranskat)abstract
- Life oil earth was formed during billions of years, exposed to, and shaped by the original physical forces such as gravitation, cosmic irradiation, atmospheric electric fields and the terrestrial magnetism. The Schumann resonances at 7.4 Hz are all example of oscillations possibly important for life.(1)) The existing organisms are created to function in harmony with these forces. However, in the late 19th century mankind introduced the use of electricity, in the early 20th century long-wave radio and in the 1940-ies short-wave radio. High frequency RF was introduced in the 50-ies as FM and television and during the very last decades, microwaves of the modern communication society spread around the world. Today, however, one third of the world's population is owner of the microwave-producing mobile phones and an even larger number is exposed to the cordless RF emitting systems. To what; extent are all living organisms affected by these, almost everywhere present radio frequency fields? And what will be the effects of many years of continuing exposure? Since 1989 Our group has studied the effects upon the mammalian blood-brain barrier (BBB) in rats by non-thermal radio frequency electromagnetic fields (RF-EMF). These have been shown to cause significantly increased leak-age of the rats' own blood albumin through the BBB of exposed rats, at energy levels of 1W/kg and below, as compared to non-exposed animals in a total series of about two thousand animals.(2)-6)) One remarkable observation is the fact that the lowest energy levels, with whole-body average power densities below 10mW/kg, give rise to the most pronounced albumin leakage. If mobile communication, even at extremely low energy levels, causes the users' own albumin to leak out through the BBB, also other unwanted and toxic molecules in the blood, may leak into the brain tissue and concentrate in and damage the neurons and glial cells of the brain. In later studies we have shown that a 2-h exposure to GSM 915 MHz, at non-thermal SAB-values of 0.2, 2 and 200 mW/kg, gives rise to significant neuronal damage, seen not only 50 days after the exposure 7) but also after 28 days but not after 14 days. Albumin extravasations and uptake into neurons was enhanced after 14 clays, but not after 28.(8)) in our continued research, also the non-thermal effects oil tissue structure and memory function of long-term exposure for 13 months are studied.(9)) We have also performed microarray analysis of brains from rats exposed to short term GSM both at 1,800 MHz and at 900MHz and have found significant effects upon gene expression of membrane associated genes as compared to control animals.(10),11)) Most of our findings support that living organisms are affected by the non-thermal radio frequency fields. Some other Studies agree while others find no effects. The mechanisms by which the EMFs may alter BBB permeability are not Well Understood. At low field strengths, the effects on body temperature are negligible and thus heating effects are not involved. A change in the physicochemical characteristics of membranes has been suggested as a cause.(12)) We have performed experiments to verify a quantum mechanical model for interaction with protein-bound ions. Our results show that controlled frequency and amplitude of ELF EM fields upon spinach plasma vesicles can steer transport over the membrane.(13)) This may be a first proof of a resonance phenomenon where appropriate levels of frequency and amplitude in the right combination have the potency to communicate with the biology of membranes and transport systems. Our study has prompted Lis to elaborate on magnetic resonance models; the Ion Cyclotron Resonance (ICR) model and the Ion Parametric Resonance (IPR) Model in an attempt to explain the occurrence of resonance frequencies. This is extensively described here under the heading: Mechanisms behind the effects of electromagnetical fields upon biology. We also bring forward the concept of solitons being active in membranes and DNA/RNA-transcription as a, possible mean to understand and prove the biological effects of EMF. The Nishinomiya-Yukawa International and Interdisciplinary Symposium 2007 raised the question: What is Life? An obvious and simple answer could be: It is DNA! The DNA strand can be looked upon as an antenna resonating in the microwave band 6GHz with its harmonics and subharmonics.(14)-18)) If this holds true, the dramatic situation might exist, that all living organisms have a receptor for the newly constructed and world-wide man-made microvaves, leading to a direct effect upon the function of DNA - in concordance with our experimental findings! Our generation invented the microwave emitters. We now have in imperative obligation to further investigate the links between EMF and biology in order to prevent possible detrimental effects of the microwaves.
|
|
| 8. |
|
|
| 9. |
- Sarve, Hamid, 1981-, et al.
(författare)
-
Registration of 2D histological images of bone implants with 3D SRuCT volumes
- 2008
-
Ingår i: Advances in visual computing. - Berlin : Springer. - 9783540896388 ; , s. 1071-1080
-
Konferensbidrag (refereegranskat)abstract
- To provide better insight in bone modeling and remodeling around implants, information is extracted using different imaging techniques. Two types of data used in this project are 2D histological images and 3D SRμCT (synchrotron radiation-based computed microtomography) volumes. To enable a direct comparison between the two modalities and to bypass the time consuming and difficult task of manual annotation of the volumes, registration of these data types is desired. In this paper, we present two 2D-3D intermodal rigid-body registration methods for the mentioned purpose. One approach is based on Simulated Annealing (SA) while the other uses Chamfer Matching (CM). Both methods use Normalized Mutual Information for measuring the correspondence between an extracted 2D-slice from the volume and the 2D histological image whereas the latter approach also takes the edge distance into account for matching the implant boundary. To speed up the process, part of the computations are done on the Graphic Processing Unit. The results show that the CM-approach provides a more reliable registration than the SA-approach. The registered slices with the CM-approach correspond visually well to the histological sections, except for cases where the implant has been damaged.
|
|
| 10. |
|
|
