Thesis (Ph. D.)--Harvard-MIT Division of Health Sciences and Technology, 2008. / Includes bibliographical references. / Evidence suggests that morphological changes of neuroanatomical structures may reflect abnormalities in neurodevelopment, or relate to a variety of disorders, such as schizophrenia and Alzheimer's disease (AD). Advances in high-resolution Magnetic Resonance Imaging (MRI) techniques allow us to study these alterations of brain structures in vivo. Previous work in studying the shape variations of brain structures has provided additional localized information compared with traditional volume-based study. However, challenges remain in finding an accurate shape presentation and conducting shape analysis with sound statistical principles. In this work, we develop methods for automatically extracting localized and multi-scale shape features and conducting statistical shape analysis of neuroanatomical structures obtained from MR images. We first develop a procedure to extract multi-scale shape features of brain structures using biorthogonal spherical wavelets. Using this wavelet-based shape representation, we build multi-scale shape models and study the localized cortical folding variations in a normal population using Principal Component Analysis (PCA). We then build a shape-based classification framework for detecting pathological changes of cortical surfaces using advanced classification methods, such as predictive Automatic Relevance Determination (pred-ARD), and demonstrate promising results in patient/control group comparison studies. Thirdly, we develop a nonlinear temporal model for studying the temporal order and regional difference of cortical folding development based on this shape representation. Furthermore, we develop a shape-guided segmentation method to improve the segmentation of sub-cortical structures, such as hippocampus, by using shape constraints obtained in the wavelet domain. / (cont.) Finally, we improve upon the proposed wavelet-based shape representation by adopting a newly developed over-complete spherical wavelet transformation and demonstrate its utility in improving the accuracy and stability of shape representations. By using these shape representations and statistical analysis methods, we have demonstrated promising results in localizing shape changes of neuroanatomical structures related to aging, neurological diseases, and neurodevelopment at multiple spatial scales. Identification of these shape changes could potentially lead to more accurate diagnoses and improved understanding of neurodevelopment and neurological diseases. / by Peng Yu. / Ph.D.

Thesis (S.M.)--Harvard-MIT Division of Health Sciences and Technology, 2011. / Vita. Cataloged from PDF version of thesis. / Includes bibliographical references (p. 91-97). / This thesis investigates the commercial viability of developing and commercializing targeted oncology drugs directed at a specific tumor mutation instead of all forms and mutations of a single target. While oncologic drugs targeted to aberrant or overexpressed pro-proliferative proteins have revolutionized cancer treatment, tumors treated for long periods may mutate over time, gain resistance to these drugs and proliferate rapidly again. I hypothesize that drugs developed to inhibit specific resistant tumor genotypes can be commercially viable from a pharmaceutical manufacturer's perspective. To assess this hypothesis empirically, I construct a patient flow model in order to quantify the treatment of CML, a relatively rare and indolent hematological malignancy with extensive clinical data available and well-delineated disease phases and response criteria. To represent the rate of diagnosis, patients are "added" to the model every month, and thereafter there is a probability that a patient may either 1) become sufficiently intolerant to his drug in order to discontinue treatment, 2) fail to respond to treatment but remain in the same disease phase, 3) fail to respond to treatment and progress to the next phase of disease, or 4) adequately respond to treatment and stay on the same drug in the same phase. Patients that fail to respond (categories 2 and 3 above) have a chance of manifesting a resistance mutation that is adequately controlled by a hypothetical drug (in addition to their current treatment) but is otherwise untreatable. The aim of this analysis is to track the number of patients that accrue the chosen resistance mutation and thus would be good candidates to receive the hypothetical drug. Patient treatment rates are converted to sales figures, and are weighed against clinical development costs, timelines, and probabilities to determine the net present value (NPV) of a project to develop the hypothetical drug. In addition, parameters are varied in order to conduct a sensitivity analysis and determine the "boundary conditions" that make a drug profitable or unprofitable. To supplement the model results and confirm the model dynamics, I interviewed investment analysts, clinical oncology thoughtleaders, academic cancer researchers and clinical, commercial and regulatory personnel from drug manufacturers to gauge their opinions on the CML market and the hurdles particular to developing drugs aimed at resistant genotypes. The conclusion I reach from this analysis is that development of a specific mutation-directed therapy for resistant CML is unlikely to be profitable. Given the significantly smaller patient population, favorable conditions in pricing and clinical development would be required to make the hypothetical candidate even marginally profitable. / by Joseph P. Sterk. / S.M.

Thesis (Ph. D.)--Harvard-MIT Division of Health Sciences and Technology, 2011. / Cataloged from PDF version of thesis. / Includes bibliographical references (p. 150-159). / This thesis builds on previous efforts to develop tactile speech-reception aids for the hearing-impaired. Whereas conventional hearing aids mainly amplify acoustic signals, tactile speech aids convert acoustic information into a form perceptible via the sense of touch. By facilitating visual speechreading and providing sensory feedback for vocal control, tactile speech aids may substantially enhance speech communication abilities in the absence of useful hearing. Research for this thesis consisted of several lines of work. First, tactual detection and temporal order discrimination by congenitally deaf adults were examined, in order to assess the practicability of encoding acoustic speech information as temporal relationships among tactual stimuli. Temporal resolution among most congenitally deaf subjects was deemed adequate for reception of tactually-encoded speech cues. Tactual offset-order discrimination thresholds substantially exceeded those measured for onset-order, underscoring fundamental differences between stimulus masking dynamics in the somatosensory and auditory systems. Next, a tactual speech transduction scheme was designed with the aim of extending the amount of articulatory information conveyed by an earlier vocoder-type tactile speech display strategy. The novel transduction scheme derives relative amplitude cues from three frequency-filtered speech bands, preserving the cross-channel timing information required for consonant voicing discriminations, while retaining low-frequency modulations that distinguish voiced and aperiodic signal components. Additionally, a sensorimotor training approach ("directed babbling") was developed with the goal of facilitating tactile speech acquisition through frequent vocal imitation of visuo-tactile speech stimuli and attention to tactual feedback from one's own vocalizations. A final study evaluated the utility of the tactile speech display in resolving ambiguities among visually presented consonants, following either standard or enhanced sensorimotor training. Profoundly deaf and normal-hearing participants trained to exploit tactually-presented acoustic information in conjunction with visual speechreading to facilitate consonant identification in the absence of semantic context. Results indicate that the present transduction scheme can enhance reception of consonant manner and voicing information and facilitate identification of syllableinitial and syllable-final consonants. The sensorimotor training strategy proved selectively advantageous for subjects demonstrating more gradual tactual speech acquisition. Simple, low-cost tactile devices may prove suitable for widespread distribution in developing countries, where hearing aids and cochlear implants remain unaffordable for most severely and profoundly deaf individuals. They have the potential to enhance verbal communication with minimal need for clinical intervention. / by Theodore M. Moallem. / Ph.D.

Thesis (Ph. D.)--Harvard-MIT Division of Health Sciences and Technology, 2006. / Includes bibliographical references. / Bilateral cochlear implantation is becoming more common as clinicians attempt to provide better sound-source localization and speech reception in noise for cochlear implant (CI) users. While some improvement over the abilities of monolateral implantees has been documented, bilateral performance for CI users is far from that achieved with normal hearing. Identifying factors that limit bilateral performance has been difficult because little is understood about CI listeners' localization abilities, their sensitivity to interaural cues, and the relationships between them. To better understand bilateral electric hearing, five bilateral CI users' abilities to locate sound sources and their sensitivities to interaural time difference (ITD) were studied in this thesis. Unlike past studies, monolateral and bilateral performance was recorded before and after exposure to daily, bilateral-CI listening using constant- and roving-level stimuli. For constant-level stimuli, increasing bilateral-listening experience improved all subjects' bilateral performance but degraded two subjects' monolateral performance. Using roving-level stimuli, increasing bilateral-listening experience also improved bilateral performance but did not alter monolateral performance. / (cont.) Our results show that depending on the method of evaluation, the benefit of bilateral CIs over monolateral CI could be overstated for some subjects. A simple decision model was used to predict subjects' localization performance based on their sensitivity to interaural time and level differences (TD and ILD) measured through their sound processors. The predicted performance indicated that the measured performance could be accounted for by subjects' ILD sensitivity but not by their ITD sensitivity alone. Poor ITD sensitivity may be one reason that bilateral CI users' localization performance is poor compared to that of normal-hearing (NH) listeners. To improve ITD sensitivity, a first step is to characterize ITD sensitivity on single, interaural electrode pairs because data in the literature is incomplete. In particular, the dependence of ITD sensitivity on the repetition rate and the number of pulses in the unmodulated pulse trains was studied. Just noticeable difference (JND) of ITD was measured with four subjects on their most ITD-sensitive, interaural electrode pair. At low rate (50 pps), ITD JND improved with increasing number of pulses, indicating integration of ongoing ITD cues. The best ITD JNDs were 85 - 354 pts. Using 800-pps trains, two subjects' ITD JND degraded with increasing number of pulses. / (cont.) Two subjects were insensitive to ITD up to 2 ms for 800-pps trains. To begin studying the impact of CI processing on ITD sensitivity, ITD JND was also measured using low-rate (50 pps) pulse trains delivered to the external input of the subjects' sound processors. ITD JND improved with increasing number of pulses. While subjects were insensitive to ongoing ITD in unmodulated, high-rate pulse trains delivered to single, interaural electrode pairs, they were sensitive to ongoing ITDs in the low-frequency modulator of high-rate pulse trains in the through-processor case. A next step toward greater understanding of bilateral electric hearing is to fully investigate the degree to which subjects are sensitive to ITD using modulated pulse trains. The results of this thesis show that there is significant localization benefit with bilateral CIs even though performance is not at the level of NL listeners. Further studies to improve ITD sensitivity may improve localization ability, which will further justify the risks and cost associated with bilateral implantation. / by Becky Bikkei Poon. / Ph.D.

Thesis (Ph.D.)--Harvard--Massachusetts Institute of Technology Division of Health Sciences and Technology, 2000. / Includes bibliographical references (leaves 135-139). / Detecting the metabolic state of atherosclerotic lesions is a promise of nuclear medicine imaging. Several researchers are developing radiopharmaceuticals for atherosclerosis imaging. In this thesis, we provided procedural guidelines to detect carotid lesions with single photon ,Emission Computed Tomography (SPECT). We first established a method to assess the requirements for "successful" lesion detection. Although this method was used to detect focal carotid lesions, it is also applicable to the detection of focal lesions in other arteries or veins. We measured lesion detectability using the output values of a 3D moving Non Pre-Whitening Matched Filter (30 mNPWMF) with the .Localization Receiver Operating Characteristics (LROC) paradigm. We simulated SPECT images of the neck using SimSPECT, our in-house analog Monte Carlo radiation transport code. We used 400 64x64 reconstructed images formed by 99mTc photons of a focal lesion in a carotid artery next to a jugular vein, both in a cylindrical water neck. We then applied the 3D \ mNPWMF along the large neck vessels. The NPWMF has been found to correlate well with human observers in simple ROC studies. We expect the mNPWMF operation to mimic a radiologist who already has a blood pool image which identifies the location of the large neck vessels. Using this detection method, we calculated that 1 to 6 kBq/cm were needed in the lesion. At large blood activity (4.6 times the surrounding tissue activity), the minimum radiopharmaceutical uptake increased by 1.6-2.9 times when the patient was lying down as opposed to sitting up. At this blood activity, a carotid dilation of 1 cm radius distracted the moving Matched Filter from lesion detection. We recommend that the blood activity be as low as possible to avoid any focal dilation from distracting our detector. We recommend that, at high blood activity, the patient be imaged in an upright position in which the jugular veins are collapsed, preventing their blood pool activity from obscuring the carotid arteries. Finally, we showed that a lesion needed 140% of the radiopharmaceutical when acquired with a radius of rotation (ROR) of25 cm instead of 15 cm. In conclusion, we assessed successfully the effect of the jugular veins and carotid dilation on detection of carotid lesions in SPECT images of the neck using the LROC detection paradigm. / by Marie-José Bélanger. / Ph.D.

Thesis: Ph. D. in Medical Engineering and Medical Physics, Harvard-MIT Program in Health Sciences and Technology, 2017. / Cataloged from PDF version of thesis. / Includes bibliographical references (pages 124-135). / The complex, dynamic, and responsive behavior of cells arises from integrated signaling pathways and regulatory networks. With advancement in our ability to engineer mammalian cells, we harness a novel set of molecular tools to develop synthetic biology-enabled applications that help facilitate our understanding of complex biological networks and cellular behaviors. The recent discovery of the Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR)-associated (Cas) system from prokaryotic adaptive immune system demonstrated unprecedented genome editing efficiency and programmability in sequence specific genome editing of mammalian cells. In this thesis, I utilized the CRISPR-Cas system to construct a combinatorial genetic perturbation platform that enables massively parallel high throughput screening of multiple gene elements. This technology platform allows systematically interrogation of higher-order interactions of genetic regulators. The later part of the work described the establishment of a genomically encoded cellular recorder with the ability to longitudinally track and record molecular events in live animals. This cellular recorder encodes cellular memory through the quantitative accumulation of targeted genomic mutations, that allows mapping of a dynamical set of gene regulatory events without the need for continuous cell imaging or destructive sampling. Together, we envision these sets of technology and tools will offer new insights into cellular process in disease and in health. / by Cheryl H. Cui. / Ph. D. in Medical Engineering and Medical Physics

Thesis (Ph. D.)--Harvard-MIT Division of Health Sciences and Technology, 2005. / Includes bibliographical references (p. 109-119). / Eukaryotic cells are permeated by a three-dimensional network of entangled filamentous proteins termed the cytoskeleton. Like scaffolding, the cytoskeleton provides rigidity and resistance to deformation from forces transmitted to and from the cell membrane. In order to model the mechanics of the cytoskeleton, the interaction of individual structural proteins must be established. To this end the relationship between two critical proteins of the actin cytoskeleton is examined. Actin reversibly assembles into filaments that provide cells with shape and confer to the cell its mechanical properties. Filamin is an abundant actin-binding protein that efficiently cross-links actin filaments in large-angle orientations, requiring the lowest concentration to convert actin filaments into a cohesive gel. Filamin dimers are composed of two 24-repeat domains that come together like a V with an actin-binding region at each free end. Analysis reveals that the repeat domains of filamin are more flexible than the self-association region of the dimer. These findings dispute the initial claim that filamin is a rigid molecule. This thesis investigates the binding modality of filamin to actin. The structure of filamin bound to actin was compared to filamin in solution and immunogold molecules bound along the filamin rod were used to map the 3-D organization of filamin-actin junctions. There is evidence that filamin binds to actin at more sites than filamin's two established N-terminal actin-binding sites. These features, flexible repeat extensions, multiple-site binding, and a rigid self-association region, make filamin a potent cross-linking agent. / (cont.) The long flexible extensions allow filamin molecules to sample a large volume of cytoplasm in search of an actin target. The binding length of filamin along actin filaments provides a less-flexible linkage from actin to the rigid filamin self-association region, enabling reproducible large angles. At the same time, filamin brings actin filaments into close proximity, creating tight network entanglement, while filamin's angle prevents linked actin filaments from slipping into dense bundles as they do with short cross-linkers like [alpha]-actinin, instead maintaining a strong, disperse network. Tightly constrained junctions support recent entropic and enthalpic models of the cytoskeleton. / by Christopher A. Hartemink. / Ph.D.

Thesis (Ph.D.)--Harvard--Massachusetts Institute of Technology Division of Health Sciences and Technology, 1999. / Includes bibliographical references. / by Jennifer Hartt Elisseeff. / Ph.D.

Thesis (Ph. D. in Bioinformatics and Integrative Genomics)--Harvard-MIT Program in Health Sciences and Technology, 2013. / Cataloged from PDF version of thesis. / Includes bibliographical references. / One in every 88 children is diagnosed with Autism Spectrum Disorders (ASDs), a set of neurodevelopmental conditions characterized by social impairments, communication deficits, and repetitive behavior. ASDs have a substantial genetic component, but the specific cause of most cases remains unknown. Understanding gene-environment interactions underlying ASD is essential for improving early diagnosis and identifying critical targets for intervention and prevention. Towards this goal, we surveyed adenosine-to-inosine (A-to-I) RNA editing in autistic brains. A-to-I editing is an epigenetic mechanism that fine-tunes synaptic function in response to environmental stimuli, shown to modulate complex behavior in animals. We used ultradeep sequencing to quantify A-to-I recoding of candidate synaptic genes in postmortem cerebella from individuals with ASD and neurotypical controls. We found unexpectedly wide distributions of human A-to-I editing levels, whose extremes were consistently populated by individuals with ASD. We correlated Ato- I editing with isoform usage, identified clusters of correlated sites, and examined differential editing patterns. Importantly, we found that individuals with ASD commonly use a dysfunctional form of the editing enzyme ADARB1. We next profiled small RNAs thought to regulate A-to-I editing, which originate from one of the most commonly altered loci in ASD, 15q11. Deep targeted sequencing of SNORD115 and SNORD116 transcripts enabled their high-resolution detection in human brains, and revealed a strong gender bias underlying their expression. The consistent 2-fold upregulation of 15q11 small RNAs in male vs. female cerebella could be important in delineating the role of this locus in ASD, a male dominant disorder. Overall, these studies provide an accurate population-level view of small RNA and A-to-I editing in human cerebella, and suggest that A-to-I editing of synaptic genes may be informative for assessing the epigenetic risk for autism. / by Alal Eran. / Ph.D.in Bioinformatics and Integrative Genomics

Thesis: Ph. D. in Medical Engineering and Medical Physics, Harvard-MIT Program in Health Sciences and Technology, 2016. / Cataloged from PDF version of thesis. / Includes bibliographical references (pages 137-148). / Patterning complex 3D features at the nanoscale offers potential applications for a wide range of fields from materials to medicine. While numerous methods have been developed to manipulate nanoscale materials, these methods are typically limited by their difficulty in creating arbitrary 3D patterns. Self-assembly of nucleic acids has emerged as a promising method for addressing this challenge due to the predictability and programmability of the material and its structure. While a diversity of DNA nanostructures have been designed by specifying complementarity rules between strands, creation of 3D nanostructures requires careful design of strand architecture, and patterns are often limited to a volume of 25 x 25 x 25 nm³ Here, we address the challenges in structural DNA nanotechnology by developing a modular DNA "brick" approach. These bricks are short, single-stranded oliogomers that can self-assemble in a single-pot reaction to a prescribed 3D shape. Using this modular approach, we demonstrate high efficiency in 3D design by generating 100 distinct, discrete 3D structures from a library of strands. We also created long-range ordering of channels, tunnels, and pores by growing micron-sized 3D periodic crystals made from DNA bricks. Finally, we applied this approach to control over 30,000 unique component strands to selfassemble into cuboids measuring over 100 nm in each dimension. These structures were further used to pattern highly complex cavities. Together, this work represents a simple, modular, and versatile method for 3D nanofabrication. This unique patterning capability of DNA bricks may enable development of new applications by providing a foundation for intricate and complex control of an unprecedented number of independent components. / by Luvena Le-Yun Ong. / Ph. D. in Medical Engineering and Medical Physics