| 1. |
- Tampu, Iulian Emil, et al.
(författare)
-
Deep-learning for thyroid microstructure segmentation in 2D OCT images
- 2021
-
Ingår i: Optical Coherence Tomography and Coherence Domain Optical Methods in Biomedicine XXV. - : SPIE - International Society for Optical Engineering.
-
Konferensbidrag (refereegranskat)abstract
- Optical coherence tomography (OCT) can provide exquisite details of tissue microstructure without traditional tissue sectioning, with potential diagnostic and intraoperative applications in a variety of clinical areas. In thyroid surgery, OCT could provide information to reduce the risk of damaging normal tissue. Thyroid tissue's follicular structure alters in case of various pathologies including the non-malignant ones which can be imaged using OCT. The success of deep learning for medical image analysis encourages its application on OCT thyroid images for quantitative analysis of tissue microstructure. To investigate the potential of a deep learning approach to segment the follicular structure in OCT images, a 2D U-Net was trained on b-scan OCT images acquired from ex vivo adult human thyroid samples a effected by a range of pathologies. Results on a pool of 104 annotated images showed a mean Dice score of 0.74±0.19 and 0.92±0.09 when segmenting the follicular structure and the surrounding tissue on the test dataset (n=10 images). This study shows that a deep learning approach for tissue microstructure segmentation in OCT images is possible. The achieved performance without requiring manual intervention encourages the application of a deep-learning method for real-time analysis of OCT data.
|
|
| 2. |
- Spyretos, Christoforos, 1996-, et al.
(författare)
-
Classification of Brain Tumour Tissue in Histopathology Images Using Deep Learning
- 2023
-
Konferensbidrag (övrigt vetenskapligt/konstnärligt)abstract
- Deep learning models have achieved prominent performance in digital pathology, with the potential to provide healthcare professionals with accurate decision-making assistance in their workflow. In this study, ViT and CNN models were implemented and compared for patch-level classification of four major glioblastoma tissue structures in histology images.A subset of the IvyGAP dataset (41 subjects, 123 images) was used, stain-normalised and patches of size 256x256 pixels were extracted. A per-subject split approach was applied to obtain training, validation and testing sets. Three models were implemented, a ViT and a CNN trained from scratch, and a ViT pre-trained on a different brain tumour histology dataset. The models' performance was assessed using a range of metrics, including accuracy and Matthew's correlation coefficient (MCC). In addition, calibration experiments were conducted and evaluated to align the models with the ground truth, utilising the temperature scaling technique. The models' uncertainty was estimated using the Monte Carlo dropout method. Lastly, the models were compared using the Wilcoxon signed-rank statistical significance test with Bonferroni correction.Among the models, the scratch-trained ViT obtained the highest test accuracy of 67% and an MCC of 0.45. The scratch-trained CNN reached a test accuracy of 49% and an MCC of 0.15, and the pre-trained ViT only achieved a test accuracy of 28% and an MCC of 0.034. Comparing the reliability graphs and metrics before and after applying temperature scaling, the subsequent experiments proceeded with the uncalibrated ViTs and the calibrated CNN. The calibrated CNN demonstrated moderate to high uncertainty across classes, and the ViTs had an overall high uncertainty. Statistically, there was no difference among the models at a significance level of 0.017. In conclusion, the scratch-trained ViT model considerably outperformed the scratch-trained CNN and the pre-trained ViT in classification. However, there was no statistically significant difference among the models.
|
|
| 3. |
- Tampu, Iulian Emil, et al.
(författare)
-
Diseased thyroid tissue classification in OCT images using deep learning: towards surgical decision support
- 2023
-
Ingår i: Journal of Biophotonics. - : Wiley. - 1864-063X .- 1864-0648. ; 16:2
-
Tidskriftsartikel (refereegranskat)abstract
- Intraoperative guidance tools for thyroid surgery based on optical coherence tomography (OCT) could aid distinguish between normal and diseased tissue. However, OCT images are difficult to interpret, thus, real-time automatic analysis could support the clinical decision-making. In this study, several deep learning models were investigated for thyroid disease classification on 2D and 3D OCT data obtained from ex vivo specimens of 22 patients undergoing surgery and diagnosed with several thyroid pathologies. Additionally, two open-access datasets were used to evaluate the custom models. On the thyroid dataset, the best performance was achieved by the 3D vision transformer model with a Matthews correlation coefficient (MCC) of 0.79 (accuracy = 0.90) for the normal-versus-abnormal classification. On the open-access datasets, the custom models achieved the best performance (MCC > 0.88, accuracy > 0.96). Results obtained for the normal-versus-abnormal classification suggest OCT, complemented with deep learning-based analysis, as a tool for real-time automatic diseased tissue identification in thyroid surgery.
|
|
| 4. |
|
|
| 5. |
|
|
| 6. |
- Haj-Hosseini, Neda, et al.
(författare)
-
Detection of brain tumor using fluorescence and optical coherence tomography
- 2015
-
Konferensbidrag (refereegranskat)abstract
- Resection of brain tumor is a challenging task as the tumor does not have clear borders and the malignant types specifically have often a diffuse and infiltrative pattern of growth. We have previously implemented and evaluated a fluorescence spectroscopy based handheld probe for detecting the 5-aminolevulinic acid induced protoporphyrin IX (PpIX) in the gliomas. To add another dimension to the brain tumor detection and volumetric analysis of the tissue that exhibits fluorescence, optical coherence tomography was investigated on tumor specimens.Material and Methods:A fluorescence microscopy and a spectroscopy system as reported previously were used for detecting the fluorescence signals [1, 2]. A total of 50 patients have been included for intraoperative assessment of the tumor borders using the fluorescence techniques. A spectral domain OCT imaging system (TELESTO II, Thorlabs, Inc., NJ, USA) with central wavelength of 1325 nm was used to study the tissue microstructure post operatively. The system has a resolution of 13 and 5.5 μm in the lateral and axial directions, respectively. Tissue specimens from three patients undergoing brain tumor surgery were studied using the OCT system.Results and Conclusion:Using fluorescence spectroscopy the tumor could be detected with a sensitivity of 0.84 which was significantly higher than that of the surgical microscope (0.30). Brain tissue appeared rather homogeneous in the OCT images however the highly malignant tissue showed a clear structural difference from the non-malignant or low malignant brain tumor tissue which could be related to the fluorescence signal intensities.
|
|
| 7. |
- Haj-Hosseini, Neda, 1980-, et al.
(författare)
-
Early Detection of Oral Potentially Malignant Disorders: A Review on Prospective Screening Methods with Regard to Global Challenges
- 2024
-
Ingår i: Journal of Maxillofacial & Oral Surgery. - New Delhi, India : Springer Science and Business Media LLC. - 0972-8279 .- 0974-942X. ; 23:1, s. 23-32
-
Tidskriftsartikel (refereegranskat)abstract
- Oral cancer is a cancer type that is widely prevalent in low-and middle-income countries with a high mortality rate, and poor quality of life for patients after treatment. Early treatment of cancer increases patient survival, improves quality of life and results in less morbidity and a better prognosis. To reach this goal, early detection of malignancies using technologies that can be used in remote and low resource areas is desirable. Such technologies should be affordable, accurate, and easy to use and interpret. This review surveys different technologies that have the potentials of implementation in primary health and general dental practice, considering global perspectives and with a focus on the population in India, where oral cancer is highly prevalent. The technologies reviewed include both sample-based methods, such as saliva and blood analysis and brush biopsy, and more direct screening of the oral cavity including fluorescence, Raman techniques, and optical coherence tomography. Digitalisation, followed by automated artificial intelligence based analysis, are key elements in facilitating wide access to these technologies, to non-specialist personnel and in rural areas, increasing quality and objectivity of the analysis while simultaneously reducing the labour and need for highly trained specialists.
|
|
| 8. |
- Haj-Hosseini, Neda, et al.
(författare)
-
Fluorescence spectroscopy and optical coherence tomography for brain tumor detection
- 2016
-
Konferensbidrag (refereegranskat)abstract
- Resection of brain tumor is a challenging task as the tumor does not have clear borders and the malignant types specifically have often a diffuse and infiltrative pattern of growth. Recently, neurosurgical microscopes have been modified to incorporate fluorescence modules for detection of tumor when 5-aminolevulinic acid (5-ALA) is used as a contrast. We have in combination with the fluorescence microscopes implemented and evaluated a fluorescence spectroscopy based handheld probe for detecting the 5-aminolevulinic acid (ALA) induced protoporphyrin IX (PpIX) in the gliomas in 50 patients intraoperatively. The results show a significantly high sensitivity for differentiating tumor from the healthy tissue and distinguished fluorescence intensity levels in the tumor cell infiltration zone around the tumor. However, knowledge on association of the quantified fluorescence signals specifically in the intermediate inflammatory zone with the infiltrative tumor cells can be complemented with volumetric tissue imaging and a higher precision histopathological analysis. In this work, a spectral domain optical coherence tomography (OCT) system with central wavelength of 1325nm has been used to image the tissue volume that the fluorescence is collected from and is evaluated against histopathological analysis for a higher precision slicing. The results show that although healthy brain has a homogenous microstructure in the OCT images, the brain tumor shows a distinguished texture in the images correlated with the PpIX fluorescence intensity and histopathology.
|
|
| 9. |
- Haj-Hosseini, Neda, et al.
(författare)
-
Low dose 5-aminolevulinic acid: Implications in spectroscopic measurements during brain tumor surgery
- 2015
-
Ingår i: Photodiagnosis and Photodynamic Therapy. - : Elsevier. - 1572-1000 .- 1873-1597. ; 12:2, s. 209-214
-
Tidskriftsartikel (refereegranskat)abstract
- BackgroundUsing 5-aminolevulinic acid (ALA) as an intraoperative fluorescence contrast has been proven to improve the resection of glioblastoma and contribute to prolonged patient survival. ALA accumulates as protoporphyrin IX (PpIX) in the tumor cells and is administered in an advised dose of 20 mg/kg body weight (b.w.) for brain tumor resection using fluorescence surgical microscopes. PpIX fluorescence availability and intensities of a four folds lower ALA dose (5 mg/kg b.w.) has been investigated in glioblastomas and skin using a spectroscopy system adapted for surgical guidance.MethodsA total of 30 adult patients diagnosed with high grade gliomas were included in the analysis. ALA was orally administered in doses of 5 mg/kg b.w. (n = 15) dissolved in orange juice or 20 mg/kg b.w. (n = 15) dissolved in water. A fluorescence spectroscopy system with a handheld fiber-optical probe was used for performing the quantitative fluorescence measurements.ResultsThe binominal comparison of the diagnostic performance parameters showed no significant statistical difference (p > 0.05). The median fluorescence values in tumor were 2-3 times higher for the high ALA dose group. No PpIX was detected in the skin of the patients in the low dose group (0/4) while PpIX was detected in the skin of the majority of the patients in the high ALA dose group (13/14).ConclusionsApplication of 5 mg/kg ALA was evaluated as equally reliable as the higher dose regarding the diagnostic performance when guidance was performed using a spectroscopic system. Moreover, no PpIX was detected in the skin of the patients.
|
|
| 10. |
|
|
