Browsing by Subject "Personalized medicine"
Now showing 1 - 6 of 6
Results Per Page
Sort Options
Item Open Access Diabetes management VIA gaussian process bandits(Bilkent University, 2021-10) Çelik, Ahmet AlparslanManagement of chronic diseases such as diabetes mellitus requires adaptation of treatment regimes based on patient characteristics and response. There is no single treatment that fits all patients in all contexts; moreover, the set of admissible treatments usually varies over the course of the disease. In this thesis, we address the problem of optimizing treatment regimes under time-varying constraints by using volatile contextual Gaussian process bandits. In particular, we propose a variant of GP-UCB with volatile arms, which takes into account the patient’s context together with the set of admissible treatments when recommending new treatments. Our Bayesian approach is able to provide treatment recommendations to the patients along with confidence scores which can be used for risk assessment. We use our algorithm to recommend bolus insulin doses for type 1 diabetes mellitus patients. We test our algorithm on in-silico subjects that come with open source implementation of the FDA-approved UVa/Padova type 1 diabetes mellitus simulator. We also compare its performance against a clinician. Moreover, we present a pilot study with a few clinicians and patients, where we design interfaces that they can interact with the model. Meanwhile, we address issues regarding privacy, safety, and ethics. Simulation studies show that our algorithm compares favorably with traditional blood glucose regulation methods.Item Open Access Discovery of cancer-specific and independent prognostic gene subsets of the slit-robo family using TCGA-PANCAN datasets(Mary Ann Liebert, 2021-12-08) Özhan, Ayşe; Tombaz, Melike; Konu, ÖzlenThe Slit-Robo family of axon guidance molecules works in concert, playing important roles in organ devel opment and cancer. Expressions of individual Slit-Robo genes have been used in calculating univariable hazard ratios (HRuni) for predicting cancer prognosis in the literature. However, Slit-Robo members do not act in dependently; hence, hazard ratios from multivariable Cox regression (HRmulti) on the whole gene set can further lead to identification of cancer-specific, novel, and independent prognostic gene pairs or modules. Herein, we obtained mRNA expressions of the Slit-Robo family consisting of four Robos (ROBO1/2/3/4) and three Slits (SLIT1/2/3), along with four types of survival outcome across cancers found in the Cancer Genome Atlas (TCGA). We used cluster heat maps to visualize closely associated pairs/modules of prognostic genes across 33 different cancers. We found a smaller number of significant genes in HRmulti than in HRuni, suggesting that the former analysis was less redundant. High ROBO4 expression emerged as relatively protective within the family, in both types of HR analyses. Multivariable Cox regression, on the other hand, revealed significantly more HR signatures containing Slit-Robo pairs acting in opposing directions than those containing Slit-Slit or Robo-Robo pairs for disease-specific survival. Furthermore, we discovered, through the online app SmulTCan’s lasso regression, Slit-Robo gene subsets that significantly differentiated between high- versus low-risk prog nosis patient groups, particularly for renal cancers and low-grade glioma. The statistical pipeline reported herein can help test independent and significant pairs/modules within a codependent gene family for cancer prog nostication, and thus should also prove useful in personalized/precision medicine research.Item Open Access Exploiting relevance for online decision-making in high-dimensions(IEEE, 2020) Turgay, Eralp; Bulucu, Cem; Tekin, CemMany sequential decision-making tasks require choosing at each decision step the right action out of the vast set of possibilities by extracting actionable intelligence from high-dimensional data streams. Most of the times, the high-dimensionality of actions and data makes learning of the optimal actions by traditional learning methods impracticable. In this work, we investigate how to discover and leverage sparsity in actions and data to enable fast learning. As our learning model, we consider a structured contextual multi-armed bandit (CMAB) with high-dimensional arm (action) and context (data) sets, where the rewards depend only on a few relevant dimensions of the joint context-arm set, possibly in a non-linear way. We depart from the prior work by assuming a high-dimensional, continuum set of arms, and allow relevant context dimensions to vary for each arm. We propose a new online learning algorithm called CMAB with Relevance Learning (CMAB-RL). CMAB-RL enjoys a substantially improved regret bound compared to classical CMAB algorithms whose regrets depend on the number of dimensions dx and da of the context and arm sets. Importantly, we show that when the learner has prior knowledge on sparsity, given in terms of upper bounds d¯¯¯x and d¯¯¯a on the number of relevant context and arm dimensions, then CMAB-RL achieves O~(T1−1/(2+2d¯¯¯x+d¯¯¯a)) regret. Finally, we illustrate how CMAB algorithms can be used for optimal personalized blood glucose control in type 1 diabetes mellitus patients, and show that CMAB-RL outperforms other contextual MAB algorithms in this task.Item Open Access Fast and accurate mapping of complete genomics reads(Academic Press, 2015) Lee, D.; Hormozdiari, F.; Xin, H.; Hach, F.; Mutlu, O.; Alkan C.Many recent advances in genomics and the expectations of personalized medicine are made possible thanks to power of high throughput sequencing (HTS) in sequencing large collections of human genomes. There are tens of different sequencing technologies currently available, and each HTS platform have different strengths and biases. This diversity both makes it possible to use different technologies to correct for shortcomings; but also requires to develop different algorithms for each platform due to the differences in data types and error models. The first problem to tackle in analyzing HTS data for resequencing applications is the read mapping stage, where many tools have been developed for the most popular HTS methods, but publicly available and open source aligners are still lacking for the Complete Genomics (CG) platform. Unfortunately, Burrows-Wheeler based methods are not practical for CG data due to the gapped nature of the reads generated by this method. Here we provide a sensitive read mapper (sirFAST) for the CG technology based on the seed-and-extend paradigm that can quickly map CG reads to a reference genome. We evaluate the performance and accuracy of sirFAST using both simulated and publicly available real data sets, showing high precision and recall rates.Item Open Access Personalizing treatments via contextual multi-armed bandits by identifying relevance(Bilkent University, 2019-08) Bulucu, CemPersonalized medicine offers specialized treatment options for individuals which is vital as every patient is different. One-size-fits-all approaches are often not effective and most patients require personalized care when dealing with various diseases like cancer, heart diseases or diabetes. As vast amounts of data became available in medicine (and otherfields including web-based recommender systems and intelligent radio networks), online learning approaches are gaining popularity due to their ability to learn fast in uncertain environments. Contextual multi-armed bandit algorithms provide reliable sequential decision-making options in such applications. In medical settings (also in other aforementioned settings), data (contexts) and actions (arms) are often high-dimensional and performances of traditional contextual multi-armed bandit approaches are almost as bad as random selection, due to the curse of dimensionality. Fortunately, in many cases the information relevant to the decision-making task does not depend on all dimensions but rather depends on a small subset of dimensions, called the relevant dimensions. In this thesis, we aim to provide personalized treatments for patients sequentially arriving over time by using contextual multi-armed bandit approaches when the expected rewards related to patient outcomes only vary on a small subset of context and arm dimensions. For this purpose,first we make use of the contextual multi-armed bandit with relevance learning (CMAB-RL) algorithm which learns the relevance by employing a novel partitioning strategy on the context-arm space and forming a set of candidate relevant dimension tuples. In this model, the set of relevant patient traits are allowed to be different for different bolus insulin dosages. Next, we consider an environment where the expected reward function defined over the context-arm space is sampled from a Gaussian process. For this setting, we propose an extension to the contextual Gaussian process upper confidence bound (CGP-UCB) algorithm, called CGP-UCB with relevance learning (CGP-UCB-RL), that learns the relevance by integrating kernels that allow weights to be associated with each dimension and optimizing the negative log marginal likelihood. Then, we investigate the suitability of this approach in the blood glucose regulation problem. Aside from applying both algorithms to the bolus insulin administration problem, we also evaluate their performance in synthetically generated environments as benchmarks.Item Open Access Whole genome sequencing: revolutionary medicine or privacy nightmare?(Institute of Electrical and Electronics Engineers, 2015) Ayday, E.; Cristofaro, Emiliano De; Hubaux, Jean-Pierre; Tsudik, G.Whole genome sequencing will soon become affordable for many individuals, but thorny privacy and ethical issues could jeopardize its popularity and thwart the large-scale adoption of genomics in healthcare and slow potential medical advances. The Web extra at http://youtu.be/As3J9NYsbbY is an audio recording of Alf Weaver interviewing Bradley Malin and Jacques Fellay about the possibilities and challenges of whole genome sequencing. © 1970-2012 IEEE.