XII tarptautinėje St Gallen krūties vėžio konferencijoje (2011) Ekspertų komisijos priimta nauja pacientų klasifikacija sisteminei terapijai atlikti, paremta biologiniais krūties vėžio subtipais, kurie apibrėžiami imunohistocheminiu tyrimu (IHC). Tačiau ši nauja navikų klasifikacija iš esmės pagrįsta pusiau kiekybiniu biožymenų raiškos vertinimu, todėl išlieka aktuali ribinių verčių nustatymo problematika. Esminiai pokyčiai IHC tyrimų srityje galimi atsiradus skaitmeninio vaizdinimo technologijoms, leidžiančiomis IHC tyrimų rezultatus analizuoti kiekybiniais parametrais. Darbe naudojant skaitmeninį vaizdo analizės metodą atliktas išsamus biologinių žymenų tyrimas leido palyginti svarbių, tačiau nepakankamai ištirtų (p53, AR, p16, BCL2, SATB1, HIF1) IHC žymenų informatyvumą su esamų prognozinių žymenų (ER, PR, HER2, Ki67) rodikliais. Ištirtas platus genetinių ir epigenetinių krūties vėžio žymenų spektras. Pirmą kartą 10 IHC žymenų rinkinio, įvertinto skaitmeninės analizės būdu, duomenys panaudoti faktorinės analizės metodu nustatyti jų variacijų vidinius veiksnius, atskleidžiančius biologinius dėsningumus ir IHC žymenų bei jų derinių informatyvumą. Šios analizės rezultatai leido naujai įvertinti publikuotų krūties vėžio IHC žymenų bei jų derinių (p16, SATB1, HIF1, Ki67/BCL2 ir kt.) informatyvumą. / The 12th International St Gallen conference on breast cancer (2011) proposed patient categorization for systemic therapy, based on intrinsic breast cancer subtypes, defined by imunohistochemistry (IHC) test results. Since this classification is based on semi-quantitative expression of IHC biomarker expression, an issue of defining and applying cutoff values remains. Essential improvement in the IHC testing has become possible with digital image analysis tools enabling quantitative evaluation of the IHC data. This study explores data obtained by digital image analysis methods applied to evaluate a comprehensive biomarker dataset (p53, AR, p16, BCL2, SATB1, HIF1) along with well established (ER, PR, HER2, Ki67) biomarkers. Also, an extensive set of genetic and epigenetic biomarkers has been tested. For the first time, the dataset of 10 IHC biomarkers, evaluated by digital analysis was explored by the means of factor analysis to establish the intrinsic factors of biological variation and informative value of IHC biomarkers and their combinatiions. The results also provided insights into the significance and combinatorial effects of the established and relatively new biomarkers (p16, SATB1, HIF1, Ki67/BCL2, etc.).

Given the strong association between aberrant nuclear morphology and tumor progression, changes in nuclear structure have remained the gold standard for cancer diagnosis for over 150 years. Recently, the rapid development of imaging hardware and computation power creates the opportunity for automated computer-aided diagnosis (CAD). Developing a robust and reliable pattern recognition pipeline is a pressing need to mine and analyze tons of nuclei data being captured. Among the rich studies on pattern recognition problems in pathology, automated nuclei detection, segmentation and cancer detection are the recurring tasks due to the importance and challenges of nuclei analysis. In this thesis, we propose and investigate the state-of-art methods in the CAD modules for maximizing the overall amount of information from images for decision making. We focus on nuclei segmentation and patient cancer detection in the nuclei image analysis pipeline. As the first step in nuclei analysis, we develop an unsupervised nuclei detection and segmentation approach for pathology images. Different from many supervised segmentation methods whose performances rely on the quality and quantity of training samples, the proposed method is able to automatically search for the nucleus contour by solving the shortest path problem with little user effort. We consider the cancer detection task as a set classification problem and propose a highly discriminative predictive model in the sense that it not only optimizes the classifier decision boundary but also transfers discriminative information to set representation learning. The innovation of the model is the integration of set representation learning and classifier training into one objective function for boosting the cancer detection performance. Experimental results showed that the new model provides significant improvements compared with state-of-art methods in the diagnostic challenges. In addition, we showed that the predictive model enables visual interpretation of discriminative nuclear characteristics representing the whole nuclei set. We believe the proposed model is quite general and provide experimental validations in several extended pattern recognition problems.

Whole Slide Images (WSIs) are digital scans containing rich pathology information. There are many available WSI datasets that can be used for a wide range of purposes such as diagnostic tasks and analysis, but the availability of labeled WSI datasets is very limited since the annotation process is both very costly and time consuming. Self-supervised learning is a way of training neural networks to learn and predict the underlying structure of input data without any labels.  AstraZeneca have developed a self-supervised learning feature extractor, the Drug-development Image Model Embeddings (DIME) pipeline, that trains on unlabeled WSIs and produces numerical embedding representations of WSI. This thesis applies the DIME-embeddings to a binary tumour classification task on the annotated Camelyon16 dataset by using the DIME pipeline as a feature extractor and train a simple binary classifier on the embedding representations instead of the WSI patches. The results are then compared to previous fully-supervised learning approaches to see if the embedding features generated by the DIME pipeline are sufficiently predictive with simple classifiers for the downstream task of binary classification.  The DIME embeddings were trained using Logistic Regression, Multi-Layer Perceptron and Gradient Boosting and the best performing model, a Multi-Layer Perceptron neural network trained on the DIME embeddings produced with an inpainting algorithm achieved a patch-level classification accuracy of 97.3%. This is very competitive results to the fully-supervised algorithms trained on the Camelyon16 dataset, beating some of them, while having a 1.1% gap to the best performing fully-supervised model. In addition to this, the performance of the DIME embeddings on reduced training sets also shows that the features captured in the DIME embeddings are sufficient.

No description available.

Artificial neural networks are networks made up of thousands and sometimes millions or more nodes also referred to as neurons. Due to the sheer scale of a network, the task of training the network can become very compute-intensive. This is because all samples need to be evaluated through the network during training, and the gradients need to be updated based on each sample`s loss. Like humans, neural networks find some samples more difficult to interpret correctly than others. By feeding the network with more difficult samples while avoiding samples it has already mastered the training process can be executed more efficiently. In the medical field neural networks are among other use cases used to identify malignant cancer in tissue samples. In such a use case being able to increase the performance of a model by 1-2 percentage units could have a huge impact on saving lives by correctly discovering malignant cancer. In this thesis project, different importance sampling methods are evaluated and tested on multiple networks and datasets. The results show how importance sampling can be utilized to faster reach a higher accuracy and save time. Not only are different importance sampling methods evaluated but also different thresholds and methods to determine when to start the importance sampling. / <p>Examensarbetet är utfört vid Institutionen för teknik och naturvetenskap (ITN) vid Tekniska fakulteten, Linköpings universitet</p>

No description available.

Deep learning (DL) has demonstrated outstanding performance in a variety of applications. With the assistance of DL, healthcare seeks to reduce labor costs and increase access to high-quality medical resources. To ensure the stability and robustness of DL applications in medicine, it is essential to estimate the uncertainty. In this thesis, the research focuses on generating an uncertainty-aware nuclei detection framework for digital pathology. A neural network (NN) with uncertainty estimation is implemented using a Bayesian deep learning method based on MC Dropout to evaluate and study the method's reliability. By evaluating and discussing the uncertainty in DL, it is possible to comprehend why it is essential to include a mechanism for measuring uncertainty. With the implementation of the framework, the results demonstrate that uncertainty-aware DL approaches enable doctors to minimize manual labeling tasks and make better decisions based on uncertainty in diagnosis and treatment. We evaluate the models in terms of both model performance and model calibration. The results demonstrate that our solution increases precision and f1 score by 15% and 11%, respectively. Using our method, the negative log likelihood (NLL) was reduced by 12% as well.

No description available.

In the field of digital histopathology, computeraideddiagnosis of digitized tissue samples with computationalalgorithms is a rising research field. The tissue samples in thisstudy are stained using chemicals that enhance the recognizabilityof different tissue structures. This staining can be highly variable,which has an impact on the performance of the computationalalgorithms. The aim of this project is to assess the use of threecolor normalization algorithms as a pre-processing step on the KIdataset from a collaborative research project between KarolinskaInstitutet and KTH Royal Institute of Technology. The colornormalization algorithms aim to reduce the color variability ofthe data. The basis of the study is an implementation of theEfficentNet Convolutional Neural Network classification model,that was adapted for the specific needs of the study. Performancewas assessed by firstly applying the color normalization filters tothe dataset and training multiple models on each of the filtereddatasets. The results from the individually trained models andthe combined results with ensemble learning techniques werethen analyzed. Our conclusions are clear, stain normalizationfilters significantly impacts classification performance metrics.The impact depends on the staining qualities of the filters.Ensemble learning techniques present a more robust performancethan the individual filters with a performance comparable to thebest performing filter. / Datorstödd diagnos av digitaliserade vävnadsprov med hjälp av beräkningsalgoritmer inom digital histopatologi är ett aktivt forskningsfält. Vävnadsproven i denna studie har färgats med kemikalier som förbättrar igenkännandet av olika vävnadsstrukturer. Kvaliteten på denna färgningsprocess kan variera, vilket har en inverkan på beräkningsalgoritmernas prestanda. Syftet med detta projekt är att utvärdera användningen av tre färgnormaliseringsalgoritmer som ett förbehandlingssteg på ett dataset från ett samarbetsprojekt mellan Karolinska Institutet och Kungliga Tekniska Högskolan. De använda färgnormaliseringsalgoritmerna syftar till att minska färgvariabiliteten i datan. Grund för studien är en implementering av klassificeringsmodellen EfficentNet, som anpassades utifrån studiens specifika behov. Prestandan bedömdes genom att först använda varje färgnormaliseringsalgoritm på datasetet och träna flera modeller på var och en av de filtrerade dataseten. Därefter analyserades resultaten från de individuella modellerna och de kombinerade resultaten med ”ensemble learning”-tekniker. Våra slutsatser är tydliga, färgnormaliseringen påverkar significant prestandamätvärdena. Dess inverkan beror på filtrens färgningsegenskaper. ”Ensemble learning” teknikerna ger en mer robust prestanda än de enskilt tränade modellerna som lika bra som det bäst presterande filtret. / Kandidatexjobb i elektroteknik 2021, KTH, Stockholm

abstract: Rapid intraoperative diagnosis of brain tumors is of great importance for planning treatment and guiding the surgeon about the extent of resection. Currently, the standard for the preliminary intraoperative tissue analysis is frozen section biopsy that has major limitations such as tissue freezing and cutting artifacts, sampling errors, lack of immediate interaction between the pathologist and the surgeon, and time consuming. Handheld, portable confocal laser endomicroscopy (CLE) is being explored in neurosurgery for its ability to image histopathological features of tissue at cellular resolution in real time during brain tumor surgery. Over the course of examination of the surgical tumor resection, hundreds to thousands of images may be collected. The high number of images requires significant time and storage load for subsequent reviewing, which motivated several research groups to employ deep convolutional neural networks (DCNNs) to improve its utility during surgery. DCNNs have proven to be useful in natural and medical image analysis tasks such as classification, object detection, and image segmentation. This thesis proposes using DCNNs for analyzing CLE images of brain tumors. Particularly, it explores the practicality of DCNNs in three main tasks. First, off-the shelf DCNNs were used to classify images into diagnostic and non-diagnostic. Further experiments showed that both ensemble modeling and transfer learning improved the classifier’s accuracy in evaluating the diagnostic quality of new images at test stage. Second, a weakly-supervised learning pipeline was developed for localizing key features of diagnostic CLE images from gliomas. Third, image style transfer was used to improve the diagnostic quality of CLE images from glioma tumors by transforming the histology patterns in CLE images of fluorescein sodium-stained tissue into the ones in conventional hematoxylin and eosin-stained tissue slides. These studies suggest that DCNNs are opted for analysis of CLE images. They may assist surgeons in sorting out the non-diagnostic images, highlighting the key regions and enhancing their appearance through pattern transformation in real time. With recent advances in deep learning such as generative adversarial networks and semi-supervised learning, new research directions need to be followed to discover more promises of DCNNs in CLE image analysis. / Dissertation/Thesis / Doctoral Dissertation Neuroscience 2019