| 1. |
|
|
| 2. |
|
|
| 3. |
|
|
| 4. |
|
|
| 5. |
|
|
| 6. |
|
|
| 7. |
- Bychkov, D, et al.
(författare)
-
Deep learning based tissue analysis predicts outcome in colorectal cancer
- 2018
-
Ingår i: Scientific reports. - : Springer Science and Business Media LLC. - 2045-2322. ; 8:1, s. 3395-
-
Tidskriftsartikel (refereegranskat)abstract
- Image-based machine learning and deep learning in particular has recently shown expert-level accuracy in medical image classification. In this study, we combine convolutional and recurrent architectures to train a deep network to predict colorectal cancer outcome based on images of tumour tissue samples. The novelty of our approach is that we directly predict patient outcome, without any intermediate tissue classification. We evaluate a set of digitized haematoxylin-eosin-stained tumour tissue microarray (TMA) samples from 420 colorectal cancer patients with clinicopathological and outcome data available. The results show that deep learning-based outcome prediction with only small tissue areas as input outperforms (hazard ratio 2.3; CI 95% 1.79–3.03; AUC 0.69) visual histological assessment performed by human experts on both TMA spot (HR 1.67; CI 95% 1.28–2.19; AUC 0.58) and whole-slide level (HR 1.65; CI 95% 1.30–2.15; AUC 0.57) in the stratification into low- and high-risk patients. Our results suggest that state-of-the-art deep learning techniques can extract more prognostic information from the tissue morphology of colorectal cancer than an experienced human observer.
|
|
| 8. |
- Jung, Christian, et al.
(författare)
-
A comparison of very old patients admitted to intensive care unit after acute versus elective surgery or intervention
- 2019
-
Ingår i: Journal of critical care. - : W B SAUNDERS CO-ELSEVIER INC. - 0883-9441 .- 1557-8615. ; 52, s. 141-148
-
Tidskriftsartikel (refereegranskat)abstract
- Background: We aimed to evaluate differences in outcome between patients admitted to intensive care unit (ICU) after elective versus acute surgery in a multinational cohort of very old patients (80 years; VIP). Predictors of mortality, with special emphasis on frailty, were assessed.Methods: In total, 5063 VIPs were induded in this analysis, 922 were admitted after elective surgery or intervention, 4141 acutely, with 402 after acute surgery. Differences were calculated using Mann-Whitney-U test and Wilcoxon test. Univariate and multivariable logistic regression were used to assess associations with mortality.Results: Compared patients admitted after acute surgery, patients admitted after elective surgery suffered less often from frailty as defined as CFS (28% vs 46%; p < 0.001), evidenced lower SOFA scores (4 +/- 5 vs 7 +/- 7; p < 0.001). Presence of frailty (CFS >4) was associated with significantly increased mortality both in elective surgery patients (7% vs 12%; p = 0.01), in acute surgery (7% vs 12%; p = 0.02).Conclusions: VIPs admitted to ICU after elective surgery evidenced favorable outcome over patients after acute surgery even after correction for relevant confounders. Frailty might be used to guide clinicians in risk stratification in both patients admitted after elective and acute surgery.
|
|
| 9. |
- Baloch, N, et al.
(författare)
-
Perineal Wound Closure Using Biological Mesh Following Extralevator Abdominoperineal Excision
- 2019
-
Ingår i: Digestive surgery. - : S. Karger AG. - 1421-9883 .- 0253-4886. ; 36:4, s. 281-288
-
Tidskriftsartikel (refereegranskat)abstract
- <b><i>Aims:</i></b> This study aimed to describe the short-term perineal healing rates in patients with perineal reconstruction using a biological mesh following extralevator abdominoperineal excision (elAPE). <b><i>Methods:</i></b> In a retrospective, descriptive single-centre cohort study, 88 consecutive patients treated with elAPE and perineal closure using a biological mesh between January 2011 and December 2015 were reviewed. All available data from electronic hospital records was collected. Patients were followed for 1 year following surgery and perineal wound status assessed at 3 months and at 1 year. <b><i>Results:</i></b> In total, 63 patients were male and all but 8 patients were treated for primary rectal cancer. All patients but 3 had received radiotherapy prior to surgery. Multivisceral excisions were performed in 19 patients. Omentoplasty was performed in 55 patients and 3 different types of meshes were used during the study period. At 3 months, 58 patients (66%) had a healed perineum. No association was detected between patient, tumour or perioperative characteristics and perineal wound status at 3 months. At 1 year, 4 patients were deceased and among the remaining 84, the perineal wound was healed in 77 patients (92%). <b><i>Conclusion:</i></b> The use of biological meshes in perineal reconstruction following elAPE is feasible and safe, and the perineal wound is healed in the majority of the patients within 3 months.
|
|
| 10. |
|
|
| 11. |
|
|
| 12. |
|
|
| 13. |
- Tjärnberg, Andreas, et al.
(författare)
-
GeneSPIDER - gene regulatory network inference benchmarking with controlled network and data properties
- 2017
-
Ingår i: Molecular Biosystems. - : Royal Society of Chemistry (RSC). - 1742-206X .- 1742-2051. ; 13:7, s. 1304-1312
-
Tidskriftsartikel (refereegranskat)abstract
- A key question in network inference, that has not been properly answered, is what accuracy can be expected for a given biological dataset and inference method. We present GeneSPIDER - a Matlab package for tuning, running, and evaluating inference algorithms that allows independent control of network and data properties to enable data-driven benchmarking. GeneSPIDER is uniquely suited to address this question by first extracting salient properties from the experimental data and then generating simulated networks and data that closely match these properties. It enables data-driven algorithm selection, estimation of inference accuracy from biological data, and a more multifaceted benchmarking. Included are generic pipelines for the design of perturbation experiments, bootstrapping, analysis of linear dependence, sample selection, scaling of SNR, and performance evaluation. With GeneSPIDER we aim to move the goal of network inference benchmarks from simple performance measurement to a deeper understanding of how the accuracy of an algorithm is determined by different combinations of network and data properties.
|
|
