Dartmouth Institute for Health Policy and Clinical Practice
Dartmouth Digital Commons (Dartmouth College)Not a member yet
8214 research outputs found
Sort by
Building Smart Textile Sensors from Bottom-up: Chemically Precise Self Assembled Materials on Textiles as a Modular Platform for Electroanalysis
No full textIn this present work, we explore the metallization of textiles and sutures for the development of a wearable portable modular platform technology for non-invasive wound healing monitoring. Currently technology for the metallization of nonconductive materials have been mainly on a 2D surfaces which restricts the bulk metallization of textile surfaces which require specialized equipment and operators which are costly, energy ineffective, and prevents miniaturization. Metallization of nonconductive fabrics are achieved through a conventional electroless deposition of copper nanoparticle species on 3D surfaces to create electrodes. These conductive textiles and sutures were characterized by pXRD, ATR-IR, SEM-EDS, and XPS and were found to contain copper zero, cuprous oxide, and copper oxide aggregated into nanoparticles. The conductive metallized surfaces of fibers with resistance values in Ω-MΩ range provide a platform for the restructuring of these surfaces into Metal Organic Frameworks as sensing materials in the electroanalysis of disease related biomarker. NO and UA are key biomarkers in chronic wound healing process of homeostasis, inflammation, proliferation, and remodeling. Sensing these biomarkers will be crucial for the future novel technological approaches to eliminating painful traditional methods of wound monitoring. In our approach copper was the first metal chosen because Copper Hexahydroxytriphenylene, Cu3(HHTP)2, a known electrochemical sensor of NO and UA. Our data suggest that electroless deposition of copper is a great candidate for creating conductive fibers for electrochemical sensing
Targeting Aspergillus fumigatus hypoxia response pathways for novel antifungal drug development.
No full textInvasive fungal disease present are difficult to diagnose and treat and present a high mortality rate across the world with approximately 1.5 million people a year succumbing to fungal infections worldwide. Currently, there are limited antifungal classes used therapeutically, and antifungal drug resistance on the rise further exacerbates the dire need for novel therapeutics with innovative mode of actions. The filamentous fungi, Aspergillus fumigatus is the causative agent of invasive pulmonary aspergillosis (IPA). It is estimated that contemporary antifungals fail in about 50% of patients with invasive filamentous fungal infections, likely due to the biofilm mode of growth exhibited in vivo by filamentous fungi. In vitro, we have shown that as biofilms mature, the establishment of low oxygen microenvironments directly contributes to their increased antifungal resistance. The transcription factor SrbA, is essential for adaptation to low oxygen, biofilm formation, virulence, and azole resistance. Therefore, we postulated inhibition of SrbA as a potential novel therapeutic strategy to combat invasive filamentous fungal infections and azole resistance.
In this dissertation we describe a high-throughput screen to identify SrbA pathway inhibitors and further characterize the mode of action of two of the small molecule hits, MBX-7591 and MBX-7498. Both of these molecules show synergy with azoles and decrease total oleic acid content, which alters phospholipid composition. Furthermore, MBX-7591 is effective in vivo in decreasing fungal burden against A. fumigatus and R. delemar. Additionally, in this discussion we additionally studied the predicted sterol transporter ArvA in A. fumigatus and identified a putative role in the conidial transition from isotropic to polarized growth and cell wall composition
Investigating Bias in Mortgage-Rate Machine Learning Models
Get PDFBanks and fintech lenders increasingly rely on computer-aided models in lending decisions. Traditional models were interpretable: decisions were based on observable factors, such as whether a borrower\u27s credit score was above a threshold value, and explainable in terms of combinations of these factors. In contrast, modern machine learning models are opaque and non-interpretable. Their opaqueness and reliance on historical data that is the artifact of past racial discrimination means these new models risk embedding and exacerbating such discrimination, even if lenders do not intend to discriminate. We calibrate two random forest classifiers using publicly available HMDA loan data and publicly available Fannie Mae loan performance data. We use two Explainable Artificial Intelligence (XAI) models, LIME and SHAP, to characterize what features drive the decisions produced by these calibrated ML lending models. Our preliminary findings suggest a significant impact of various racial factors within a model\u27s decision-making process when it has access to such information, as seen in the model trained on HMDA data. These results highlight the need for further investigation to understand and address these influences in depth
Math, ChatGPT, and You: The Problem with Mathematical Accuracy in Large Language Models
Get PDFChatGPT and other Large Language Models (LLMs) currently do a good job at generating novel text across many domains, but math remains a consistent issue when it comes to the accuracy of answers generated by these models. My research into various ways to manipulate the model have led me to the conclusion that a general closed form solution to help LLMs with math is both unrealistic and likely impossible. LLMs can be trained more successfully as you narrow the problem space, but consideration must be taken on the part of human user to recognize when an LLM is detrimental to your solution and a traditional programming solution should be taken instead
Examining Differences in Concept Representation Across Similarity Spaces Between Humans and Large Language Models
Get PDFThe replication of human concept representation is a critical task for the pursuit of artificial general intelligence. With the recent influx of large language models that demonstrate text-generation capabilities nearly on par with humans, the question stands on whether these large language models can capture concepts within language. We examine this question by exploring differences in concept representation across similarity spaces between humans and LLMs. We find that, while concept representation within LLMs does partially mimic human concept representation, LLMs are greatly limited by their dependence on semantic information and cannot therefore develop an understanding of human social code or morality. Our results suggest that there are limitations imposed by the model design of LLMs that will prevent full replication of human concept representation
Latent Auto-recursive Composition Engine: A Generative System for Creative Expression in Human-AI Collaboration
Get PDFThis thesis investigates the shifting boundaries of art in the era of Generative AI, crit-ically examining the essence of art and the legitimacy of AI-generated works. Despitesignificant advancements in the quality and accessibility of art through generativeAI, such creations frequently encounter skepticism regarding their status as authenticart. To address this skepticism, the study explores the role of creative agency in var-ious generative AI workflows and introduces an ”artist-in-the-loop” system tailoredfor image generation models like Stable Diffusion. This system aims to deepen theartist’s engagement and understanding of the creative process. Additionally, a noveltool, the Latent Auto-recursive Composition Engine (LACE), which integrates Pho-toshop and ControlNet with Stable Diffusion, is introduced to improve transparencyand control. This approach not only broadens the scope of computational creativitybut also enhances artists’ ownership of AI-generated art, bridging the divide betweenAI-driven and traditional human artistry in the digital landscape
Nebraska, Anyone? An Avid Hiker Navigates Around Jeep Tracks and High Prairie
Get PDFAn avid adventurer tries hiking around Jeep tracks in Fort Robinson State Park and through high prairie to Toadstool Geologic Park
The Stranger on Moosilauke: A Hiker Revits the Haunting Tragedy of a Frigid Day
Get PDFA hiker recounts meeting Roy Sanford on the slopes of Mount Moosilauke in the western White Mountains of New Hampshire, later learning that he died trying to summit in bad weather
Letters
Get PDFResponses to William Geller’s Winter/Spring 2023 article on bushwhacking in the Mahoosuc range in Maine
Alpina: A Semiannual Review of Mountaineering in the Greater Ranges
Get PDFAmong the many updates on mountaineering in the greater ranges in summer and fall 2023: Samson Zebturiah Barner is arrested near Smith Rock in central Oregon for taking several weapons to a climbing event. Three climbers open a new route on Pik Alpinist in Kyrgyzstan. A brief review of the New York Times article on the discovery of Janet Johnson’s camera on Aconcagua 50 years after she and John Cooper died there. On Everest, 3,600 climbers attempted the peak in 2023; eighteen died, 100 years after the biggest Everest mystery of all: the disappearance of Mallory and Irvine. Three Americans climb Jannu. Two British climbers summit Surma-Sarovar, and two climbers top out on Yansa Tsenji. Another look at Kristin Harila’s behavior around a dying porter on K2, and a postscript about the death of Harila’s companion Tenjen Sherpa. Audrey Salkeld, a British mountaineering historian, died October 11, 2023