Deep Learning techniques are widely used across various medical imaging applications. However, they are often fine-tuned for a specific modality and are not generalizable when it comes to new modalities or datasets. One of the main reasons for this is large data variations for e.g., the dynamic range of intensity values is large across multi-modal images. The goal of the project is to develop a method to address multi-modal learning that aims at segmenting liver from Computed Tomography (CT) images and abdominal organs from Magnetic Resonance (MR) images using deep learning techniques. In this project, a self-supervised approach is adapted to attain domain adaptation across images while retaining important 3D information from medical images using a simple 3D-UNet with a few auxiliary tasks. The method comprises of two main steps: representation learning via self-supervised learning (pre-training) and fully supervised learning (fine-tuning). Pre-training is done using a 3D-UNet as a base model along with some auxiliary data augmentation tasks to learn representation through texture, geometry and appearances. The second step is fine-tuning the same network, without the auxiliary tasks, to perform the segmentation tasks on CT and MR images. The annotations of all organs are not available in both modalities. Thus the first step is used to learn general representation from both image modalities; while the second step helps to fine-tune the representations to the available annotations of each modality. Results obtained for each modality were submitted online, and one of the evaluations obtained was in the form of DICE score. The results acquired showed that the highest DICE score of 0.966 was obtained for CT liver prediction and highest DICE score of 0.7 for MRI abdominal segmentation. This project shows the potential to achieve desired results by combining both self and fully-supervised approaches.

In recent decades, various software systems have gradually become the basis of our society. Programmers search existing code snippets from time to time in their daily life. It would be beneficial and meaningful to have better solutions for the task of semantic code search, which is to find the most semantically relevant code snippets for a given query. Our approach is to introduce tree representations by multi-modal learning. The core idea is to enrich semantic information for code snippets by preparing data of different modalities, and meanwhile ignore syntactic information. We design one novel tree structure named Simplified Semantic Tree and then extract RootPath representations from that. We utilize RootPath representation to complement the conventional sequential representation, namely the token sequence of the code snippet. Our multi-modal model receives code-query pair as input and computes similarity score as output, following the pseudo-siamese architecture. For each pair, besides the ready-made code sequence and query sequence, we extra one extra tree sequence from Simplified Semantic Tree. There are three encoders in our model, and they respectively encode these three sequences as vectors of the same length. Then we combine the code vector with the tree vector for one joint vector, which is still of the same length, as the multi-modal representation for the code snippet. We introduce triplet loss to ensure vectors of code and query in the same pair be close at the shared vector space. We conduct experiments in one large-scale multi-language corpus, with comparisons of strong baseline models by specified performance metrics. Among baseline models, the simplest Neural Bag-of-Words model is with the most satisfying performance. It indicates that syntactic information is likely to distract complex models from critical semantic information. Results show that our multi-modal representation approach performs better because it surpasses baseline models by far in most cases. The key to our multi-modal model is that it is totally about semantic information, and it learns from data of multiple modalities. / Under de senaste decennierna har olika programvarusystem gradvis blivit basen i vårt samhälle. Programmerare söker i befintliga kodavsnitt från tid till annan i deras dagliga liv. Det skulle vara fördelaktigt och meningsfullt att ha bättre lösningar för uppgiften att semantisk kodsökning, vilket är att hitta de mest semantiskt relevanta kodavsnitten för en given fråga. Vår metod är att introducera trädrepresentationer genom multimodal inlärning. Grundidén är att berika semantisk information för kodavsnitt genom att förbereda data med olika modaliteter och samtidigt ignorera syntaktisk information. Vi designar en ny trädstruktur med namnet Simplified Semantic Tree och extraherar sedan RootPath-representationer från det. Vi använder RootPath-representation för att komplettera den konventionella sekvensrepresentationen, nämligen kodsekvensens symbolsekvens. Vår multimodala modell får kodfrågeställningar som inmatning och beräknar likhetspoäng som utgång efter den pseudo-siamesiska arkitekturen. För varje par, förutom den färdiga kodsekvensen och frågesekvensen, extrager vi en extra trädsekvens från Simplified Semantic Tree. Det finns tre kodare i vår modell, och de kodar respektive tre sekvenser som vektorer av samma längd. Sedan kombinerar vi kodvektorn med trädvektorn för en gemensam vektor, som fortfarande är av samma längd som den multimodala representationen för kodavsnittet. Vi introducerar tripletförlust för att säkerställa att vektorer av kod och fråga i samma par är nära det delade vektorn. Vi genomför experiment i ett storskaligt flerspråkigt korpus, med jämförelser av starka baslinjemodeller med specificerade prestandametriker. Bland baslinjemodellerna är den enklaste Neural Bag-of-Words-modellen med den mest tillfredsställande prestanda. Det indikerar att syntaktisk information sannolikt kommer att distrahera komplexa modeller från kritisk semantisk information. Resultaten visar att vår multimodala representationsmetod fungerar bättre eftersom den överträffar basmodellerna i de flesta fall. Nyckeln till vår multimodala modell är att den helt handlar om semantisk information, och den lär sig av data om flera modaliteter.