| 1. |
- Mondlane, Gracinda, 1987-
(författare)
-
Comparative study of Radiation Therapy of Targets in the Upper Abdomen with Photon- or Scanned Proton-beams
- 2017
-
Licentiatavhandling (övrigt vetenskapligt/konstnärligt)abstract
- Recently, there has been an increase in the number of proton beam therapy (PBT) centers operating worldwide. For certain cases, proton beams have been shown to provide dosimetric and radiobiological advantages when used for cancer treatment, compared to the regular photon-beam based treatments. Under ideal circumstances, the dose given to the tissues surrounding a target can be reduced with PBT. The risk for side effects following treatment is then expected to decrease. Until present, mainly stationary targets, e.g. targets in the brain, have been treated with PBT. There is currently a growing interest to treat also target volumes in other parts of the body with PBT. However, there are sources of uncertainties, which must be more carefully considered when PBT is used, especially for PBT carried out with scanned proton beams. PBT is more sensitive to anatomical changes, e.g. organ motion or a variable gas content in the intestines, which requires that special precautions are taken prior to treating new tumour sites. In photon beam radiotherapy (RT) of moving targets, the main consequence of organ motion is the loss of sharpness of the dose gradients (dose smearing). When scanned proton beams are used, dose deformation caused by the fluctuations in the proton beam range, due to varying tissue heterogeneities (e.g., the ribs moving in and out of the beam path) and the so-called interplay effect, can be expected to impact the dose distributions in addition to the dose smearing. The dosimetric uncertainties, if not accounted for, may cause the planned and accurately calculated dose distribution to be distorted, compromising the main goal of RT of achieving the maximal local disease control while accepting certain risks for normal tissue complications.Currently there is a lack of clinical follow-up data regarding the outcome of PBT for different tumour sites, in particular for extra-cranial tumour sites in moving organs. On the other hand, the use of photon beams for this kind of cancer treatment is well-stablished. A treatment planning comparison between RT carried out with photons and with protons may provide guidelines for when PBT could be more suitable. New clinical applications of particle beams in cancer therapy can also be transferred from photon-beam treatments, for which there is a vast clinical experience. The evaluation of the different uncertainties influencing RT of different tumour sites carried out with photon- and with proton-beams, will hopefully create an understanding for the feasibility of treating cancers with scanned proton beams instead of photon beams. The comparison of two distinct RT modalities is normally performed by studying the dosimetric values obtained from the dose volume histograms (DVH). However, in dosimetric evaluations, the outcome of the treatments in terms of local disease control and healthy tissue toxicity are not estimated. In this regard, radiobiological models can be an indispensable tool for the prediction of the outcome of cancer treatments performed with different types of ionising radiation. In this thesis, different factors that should be taken into consideration in PBT, for treatments influenced by organ motion and density heterogeneities, were studied and their importance quantified.This thesis consists of three published articles (Articles I, II and III). In these reports, the dosimetric and biological evaluations of photon-beam and scanned proton-beam RT were performed and the results obtained were compared. The studies were made for two tumour sites influenced by organ motion and density changes, gastric cancer (GC) and liver metastases. For the GC cases, the impact of changes in tissue density, resulting from variable gas content (which can be observed inter-fractionally), was also studied. In this thesis, both conventional fractionations (implemented in the planning for GC treatments) and hypofractionated regimens (implemented in the planning for the liver metastases cases) were considered. In this work, it was found that proton therapy provided the possibility to reduce the irradiations of the normal tissue located near the target volumes, compared to photon beam RT. However, the effects of density changes were found to be more pronounced in the plans for PBT. Furthermore, with proton beams, the reduction of the integral dose given to the OARs resulted in reduced risks of treatment-induced secondary malignancies.
|
|
| 2. |
- Astaraki, Mehdi, PhD Student, 1984-, et al.
(författare)
-
A Comparative Study of Radiomics and Deep-Learning Based Methods for Pulmonary Nodule Malignancy Prediction in Low Dose CT Images
- 2021
-
Ingår i: Frontiers in Oncology. - : Frontiers Media SA. - 2234-943X. ; 11
-
Tidskriftsartikel (refereegranskat)abstract
- Objectives: Both radiomics and deep learning methods have shown great promise in predicting lesion malignancy in various image-based oncology studies. However, it is still unclear which method to choose for a specific clinical problem given the access to the same amount of training data. In this study, we try to compare the performance of a series of carefully selected conventional radiomics methods, end-to-end deep learning models, and deep-feature based radiomics pipelines for pulmonary nodule malignancy prediction on an open database that consists of 1297 manually delineated lung nodules.Methods: Conventional radiomics analysis was conducted by extracting standard handcrafted features from target nodule images. Several end-to-end deep classifier networks, including VGG, ResNet, DenseNet, and EfficientNet were employed to identify lung nodule malignancy as well. In addition to the baseline implementations, we also investigated the importance of feature selection and class balancing, as well as separating the features learned in the nodule target region and the background/context region. By pooling the radiomics and deep features together in a hybrid feature set, we investigated the compatibility of these two sets with respect to malignancy prediction.Results: The best baseline conventional radiomics model, deep learning model, and deep-feature based radiomics model achieved AUROC values (mean ± standard deviations) of 0.792 ± 0.025, 0.801 ± 0.018, and 0.817 ± 0.032, respectively through 5-fold cross-validation analyses. However, after trying out several optimization techniques, such as feature selection and data balancing, as well as adding context features, the corresponding best radiomics, end-to-end deep learning, and deep-feature based models achieved AUROC values of 0.921 ± 0.010, 0.824 ± 0.021, and 0.936 ± 0.011, respectively. We achieved the best prediction accuracy from the hybrid feature set (AUROC: 0.938 ± 0.010).Conclusion: The end-to-end deep-learning model outperforms conventional radiomics out of the box without much fine-tuning. On the other hand, fine-tuning the models lead to significant improvements in the prediction performance where the conventional and deep-feature based radiomics models achieved comparable results. The hybrid radiomics method seems to be the most promising model for lung nodule malignancy prediction in this comparative study.
|
|
| 3. |
- Astaraki, Mehdi, PhD Student, 1984-
(författare)
-
Advanced Machine Learning Methods for Oncological Image Analysis
- 2022
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- Cancer is a major public health problem, accounting for an estimated 10 million deaths worldwide in 2020 alone. Rapid advances in the field of image acquisition and hardware development over the past three decades have resulted in the development of modern medical imaging modalities that can capture high-resolution anatomical, physiological, functional, and metabolic quantitative information from cancerous organs. Therefore, the applications of medical imaging have become increasingly crucial in the clinical routines of oncology, providing screening, diagnosis, treatment monitoring, and non/minimally-invasive evaluation of disease prognosis. The essential need for medical images, however, has resulted in the acquisition of a tremendous number of imaging scans. Considering the growing role of medical imaging data on one side and the challenges of manually examining such an abundance of data on the other side, the development of computerized tools to automatically or semi-automatically examine the image data has attracted considerable interest. Hence, a variety of machine learning tools have been developed for oncological image analysis, aiming to assist clinicians with repetitive tasks in their workflow.This thesis aims to contribute to the field of oncological image analysis by proposing new ways of quantifying tumor characteristics from medical image data. Specifically, this thesis consists of six studies, the first two of which focus on introducing novel methods for tumor segmentation. The last four studies aim to develop quantitative imaging biomarkers for cancer diagnosis and prognosis.The main objective of Study I is to develop a deep learning pipeline capable of capturing the appearance of lung pathologies, including lung tumors, and integrating this pipeline into the segmentation networks to leverage the segmentation accuracy. The proposed pipeline was tested on several comprehensive datasets, and the numerical quantifications show the superiority of the proposed prior-aware DL framework compared to the state of the art. Study II aims to address a crucial challenge faced by supervised segmentation models: dependency on the large-scale labeled dataset. In this study, an unsupervised segmentation approach is proposed based on the concept of image inpainting to segment lung and head-neck tumors in images from single and multiple modalities. The proposed autoinpainting pipeline shows great potential in synthesizing high-quality tumor-free images and outperforms a family of well-established unsupervised models in terms of segmentation accuracy.Studies III and IV aim to automatically discriminate the benign from the malignant pulmonary nodules by analyzing the low-dose computed tomography (LDCT) scans. In Study III, a dual-pathway deep classification framework is proposed to simultaneously take into account the local intra-nodule heterogeneities and the global contextual information. Study IV seeks to compare the discriminative power of a series of carefully selected conventional radiomics methods, end-to-end Deep Learning (DL) models, and deep features-based radiomics analysis on the same dataset. The numerical analyses show the potential of fusing the learned deep features into radiomic features for boosting the classification power.Study V focuses on the early assessment of lung tumor response to the applied treatments by proposing a novel feature set that can be interpreted physiologically. This feature set was employed to quantify the changes in the tumor characteristics from longitudinal PET-CT scans in order to predict the overall survival status of the patients two years after the last session of treatments. The discriminative power of the introduced imaging biomarkers was compared against the conventional radiomics, and the quantitative evaluations verified the superiority of the proposed feature set. Whereas Study V focuses on a binary survival prediction task, Study VI addresses the prediction of survival rate in patients diagnosed with lung and head-neck cancer by investigating the potential of spherical convolutional neural networks and comparing their performance against other types of features, including radiomics. While comparable results were achieved in intra-dataset analyses, the proposed spherical-based features show more predictive power in inter-dataset analyses.In summary, the six studies incorporate different imaging modalities and a wide range of image processing and machine-learning techniques in the methods developed for the quantitative assessment of tumor characteristics and contribute to the essential procedures of cancer diagnosis and prognosis.
|
|
| 4. |
- Astaraki, Mehdi, PhD Student, 1984-, et al.
(författare)
-
Benign-malignant pulmonary nodule classification in low-dose CT with convolutional features
- 2021
-
Ingår i: Physica medica (Testo stampato). - : Elsevier BV. - 1120-1797 .- 1724-191X. ; 83, s. 146-153
-
Tidskriftsartikel (refereegranskat)abstract
- Purpose: Low-Dose Computed Tomography (LDCT) is the most common imaging modality for lung cancer diagnosis. The presence of nodules in the scans does not necessarily portend lung cancer, as there is an intricate relationship between nodule characteristics and lung cancer. Therefore, benign-malignant pulmonary nodule classification at early detection is a crucial step to improve diagnosis and prolong patient survival. The aim of this study is to propose a method for predicting nodule malignancy based on deep abstract features.Methods: To efficiently capture both intra-nodule heterogeneities and contextual information of the pulmonary nodules, a dual pathway model was developed to integrate the intra-nodule characteristics with contextual attributes. The proposed approach was implemented with both supervised and unsupervised learning schemes. A random forest model was added as a second component on top of the networks to generate the classification results. The discrimination power of the model was evaluated by calculating the Area Under the Receiver Operating Characteristic Curve (AUROC) metric. Results: Experiments on 1297 manually segmented nodules show that the integration of context and target supervised deep features have a great potential for accurate prediction, resulting in a discrimination power of 0.936 in terms of AUROC, which outperformed the classification performance of the Kaggle 2017 challenge winner.Conclusion: Empirical results demonstrate that integrating nodule target and context images into a unified network improves the discrimination power, outperforming the conventional single pathway convolutional neural networks.
|
|
| 5. |
- Astaraki, Mehdi, PhD Student, 1984-, et al.
(författare)
-
Early survival prediction in non-small cell lung cancer from PET/CT images using an intra-tumor partitioning method
- 2019
-
Ingår i: Physica medica (Testo stampato). - : Elsevier BV. - 1120-1797 .- 1724-191X. ; 60, s. 58-65
-
Tidskriftsartikel (refereegranskat)abstract
- PurposeTo explore prognostic and predictive values of a novel quantitative feature set describing intra-tumor heterogeneity in patients with lung cancer treated with concurrent and sequential chemoradiotherapy.MethodsLongitudinal PET-CT images of 30 patients with non-small cell lung cancer were analysed. To describe tumor cell heterogeneity, the tumors were partitioned into one to ten concentric regions depending on their sizes, and, for each region, the change in average intensity between the two scans was calculated for PET and CT images separately to form the proposed feature set. To validate the prognostic value of the proposed method, radiomics analysis was performed and a combination of the proposed novel feature set and the classic radiomic features was evaluated. A feature selection algorithm was utilized to identify the optimal features, and a linear support vector machine was trained for the task of overall survival prediction in terms of area under the receiver operating characteristic curve (AUROC).ResultsThe proposed novel feature set was found to be prognostic and even outperformed the radiomics approach with a significant difference (AUROCSALoP = 0.90 vs. AUROCradiomic = 0.71) when feature selection was not employed, whereas with feature selection, a combination of the novel feature set and radiomics led to the highest prognostic values.ConclusionA novel feature set designed for capturing intra-tumor heterogeneity was introduced. Judging by their prognostic power, the proposed features have a promising potential for early survival prediction.
|
|
| 6. |
- Astaraki, Mehdi, PhD Student, 1984-, et al.
(författare)
-
Multimodal brain tumor segmentation with normal appearance autoencoder
- 2019
-
Ingår i: International MICCAI Brainlesion Workshop. - Cham : Springer Nature. ; , s. 316-323
-
Konferensbidrag (refereegranskat)abstract
- We propose a hybrid segmentation pipeline based on the autoencoders’ capability of anomaly detection. To this end, we, first, introduce a new augmentation technique to generate synthetic paired images. Gaining advantage from the paired images, we propose a Normal Appearance Autoencoder (NAA) that is able to remove tumors and thus reconstruct realistic-looking, tumor-free images. After estimating the regions where the abnormalities potentially exist, a segmentation network is guided toward the candidate region. We tested the proposed pipeline on the BraTS 2019 database. The preliminary results indicate that the proposed model improved the segmentation accuracy of brain tumor subregions compared to the U-Net model.
|
|
| 7. |
- Astaraki, Mehdi, PhD Student, 1984-, et al.
(författare)
-
Normal appearance autoencoder for lung cancer detection and segmentation
- 2019
-
Ingår i: International Conference on Medical Image Computing and Computer-Assisted Intervention. - Cham : Springer Nature. ; , s. 249-256
-
Konferensbidrag (refereegranskat)abstract
- One of the major differences between medical doctor training and machine learning is that doctors are trained to recognize normal/healthy anatomy first. Knowing the healthy appearance of anatomy structures helps doctors to make better judgement when some abnormality shows up in an image. In this study, we propose a normal appearance autoencoder (NAA), that removes abnormalities from a diseased image. This autoencoder is semi-automatically trained using another partial convolutional in-paint network that is trained using healthy subjects only. The output of the autoencoder is then fed to a segmentation net in addition to the original input image, i.e. the latter gets both the diseased image and a simulated healthy image where the lesion is artificially removed. By getting access to knowledge of how the abnormal region is supposed to look, we hypothesized that the segmentation network could perform better than just being shown the original slice. We tested the proposed network on the LIDC-IDRI dataset for lung cancer detection and segmentation. The preliminary results show the NAA approach improved segmentation accuracy substantially in comparison with the conventional U-Net architecture.
|
|
| 8. |
|
|
| 9. |
|
|
| 10. |
- dos Santos Matias, Lucilio
(författare)
-
Optimization of brachytherapy for cervical cancer using inverse planning algorithms.
- 2015
-
Licentiatavhandling (övrigt vetenskapligt/konstnärligt)abstract
- Carcinoma of the cervix is a global problem. Brachytherapy (BT) is one of the mainradiation therapy components used in the management of cervical cancer. With theadvent of scientific and technological developments in treatment planning, inverse optimizationin BT has been imposed; however, to harness the full potential of inverseplanning in brachytherapy, its thorough comparison with manual optimization methodsis warranted.Although inverse optimization algorithms are based on different mathematical approaches,their goals are similar. The underlying principles of these algorithms willallow them to be applied with the aim of respecting normal structures absorbed doselimits while delivering high enough tumouricidal dose.In this work, the physical parameters minimum dose received by 98% and 90% ofthe target volume represented by D98 and D90, respectively, were used to evaluate thetreatment plans with respect to the target while the minimum dose received by 2cm3volume, D2cm3 , was used to investigate complications in organs at risk (OARs). Theconformity index (COIN), was used to describe the coverage of the target by the prescribeddose (PD) and the fraction of each OAR volume that receives a critical dose,which may cause complication. The treatment plan evaluation was also performedin terms of the complication-free tumour control probability, P+. The physical andradiobiological evaluation corresponding to plans obtained by the inverse planningsimulated annealing algorithm (IPSA) and the hybrid inverse planning optimization(HIPO) have been compared with corresponding ones for plans obtained using a manualgraphical optimization method.The main observations of this work are that well tuned class solutions of inverseoptimization methods are able to produce similar dose volume histograms to thoseproduced with manual graphical optimization and inverse methods have the potentialto spare organs at risk while delivering acceptable dose to the target. In addition, radiobiologicalindexes such as the P+ can be useful complements to physical parametersin treatment plan evaluation.
|
|
