Upload your resume. PhdFit's six research agents compare your background with faculty profiles, recent publications, lab focus, and outreach opportunities, then rank professors with evidence you can review.
Resume-aware matchingEvidence from recent researchOutreach drafts included
Advisor Match Brief Resume uploaded
Resume signals5 found
NLPHCIFairnessGraph MLPyTorch
Six-agent analysisRunning
Axel · Candidate AnalystResume parsed
Nova · Professor ResearcherRecent papers
Echo · Match ExplainerEvidence linked
Atlas · Strategy AdvisorFit ranked
Lyra · Outreach WriterDraft queued
Juno · Meeting Prep CompilerNext step ready
SC
Dr. Sarah Chen
Stanford · NLP & interpretability
91
fit
Fairness + NLP projects overlap with her recent work.
Strong methods fit: model evaluation and interpretability.
Clear outreach angle for current trustworthy AI research.
Outreach angle
Ask how her lab is extending interpretability methods into fairness audits for real-world AI systems.
Matthew L. Jones is the Smith Family Professor of History at Princeton University, focusing on the history of recent information technologies, intelligence, and the history of science and technology in early modern Europe. He received his A.B. and Ph.D. from Harvard University in 1994 and 2000, respectively, and earned an M.Phil. from Cambridge. Prior to his current position, he taught at Columbia University for twenty-three years. His research includes a forthcoming book on state surveillance of communications from the 1970s to the present, with particular attention to signals intelligence agencies of the US and its allies. Jones has published two books, 'The Good Life in the Scientific Revolution' and 'Reckoning with Matter,' and co-authored 'How Data Happened,' a history of data, statistics, and machine learning from the 1800s to today. He has received fellowships from the Mellon Foundation, the Guggenheim Foundation, the Sloan Foundation, and the National Science Foundation, and is a CIFAR fellow in the Future Flourishing project.
Abstract Aim To assess compliance of management of upper limb bite injuries with Departmental guidelines, generate an action plan to address areas requiring improvement, and close the loop to review outcomes. Method A total of 38 patients in April 2023 and 27 patients in October 2023 presented to University Hospital Coventry. Trust electronic software was used to identify patients and collect data that reflected the standards set by the Departmental guideline, including treatment given at first presentation and subsequent management. Results The Departmental guideline was updated in accordance with The British Society for Surgery of the Hand. This resulted in an improved 86% of all patients, including 100% with infection, who required an operation to be completed within 24 hours. First contact practitioners provided a thorough washout in 25% of patients in the Emergency Department. An overall 30% required a formal washout in theatre by a surgeon, of which 25% required a second operation. Further, all but one patient initially discharged from the Emergency Department to Hand Trauma Clinic were managed conservatively, and 23% required more than one follow-up appointment. Additional recommendations focused on enabling better care by first contact practitioners, including clinical training for simple wound excision and washout in ED and encouraging improved documentation, as well as providing a dedicated minor ops room with accessible equipment. Conclusions A closed-loop audit on the management of upper limb bite injuries showed a marked improvement in achieving compliance with a revised Departmental guideline, with further recommendations targeted at enabling first contact practitioners.
Journal of the American College of Emergency Physicians Open · 2023-02-25 · 1 citations
articleOpen access
A 40-year-old black male with a past medical history of sickle cell disease, cardiomyopathy, and pulmonary hypertension presents as a transfer from an outside facility to a tertiary care hospital for evaluation after a formal echocardiogram reported “biatrial enlargement and small cystic structure adjacent to the right atrium.” The patient arrived at the receiving hospital in acute severe distress with altered mental status, tachycardia, tachypnea, hypoxia, and hypotension with near agonal respirations on bilevel positive airway pressure. He was also noted to have bilateral jugular venous distention to the angle of mandibles and bilateral lower extremity edema. Given the acuity of the patient's condition, a bedside point-of-care (POC) ultrasound was done using the Rapid Ultrasound for Shock and Hypotension (RUSH) protocol. Bilateral lung slide was present with a bilateral A-line profile. Inferior vena cava (IVC) interrogation showed a markedly dilated non-distensible IVC. In the apical 4-chamber cardiac view, with color Doppler, it became obvious that there was a very large extracardiac cystic mass nearly completely compressing the right atrium (Video 1). Mitral valve inflow velocity was assessed via pulse wave Doppler and was found to demonstrate echocardiographic pulsus paradoxus (Figure 1). The emergency physician made the diagnosis of an extracardiac cystic mass compressing the right atrium causing pericardial tamponade rapidly at the bedside. A multidisciplinary team was mobilized by the emergency physician and the epicardial cyst was drained under ultrasound guidance in the emergency department resulting in 1300 cc of serous fluid with the restoration of the right atrium volume, normalization of the patient's vital signs, and return to his normal mental status. Pericardial tamponade occurs when there is an impairment of ventricular filling due to atrial compression by pericardial fluid or blood.1 Far more rarely, cardiac tamponade can be caused by extrinsic compression.2 Traumatic pericardial effusions typically occur rapidly and present as acute hemodynamic compromise in the setting of trauma. Medical causes of tamponade can be much more insidious given the usually slower accumulation of pericardial fluid. Pericardial tamponade presents typically as either cardiogenic or obstructive shock. Although physical exam findings such as Beck's triad, are often cited, echocardiography provides an invaluable tool in the detection of tamponade.3 In one case series, the sensitivity of the presence of all 3 components of Becks Triad was 0%, whereas any single element had a sensitivity 50%.4 Emergency physicians can rapidly diagnose pericardial tamponade with POC ultrasound and thus render lifesaving treatment, as in this case.
Like the polar bear beleaguered by global warming, artificial intelligence (AI) serves as the charismatic megafauna of an entangled set of local and global histories of science, technology and economics. This Themes issue develops a new perspective on AI that moves beyond conventional origin myths – AI was invented at Dartmouth in the summer of 1956, or by Alan Turing in 1950 – and reframes contemporary critique by establishing plural genealogies that situate AI within deeper histories and broader geographies. ChatGPT and art produced by AI are described as generative but are better understood as forms of pastiche based upon the use of existing infrastructures, often in ways that reflect stereotypes. The power of these tools is predicated on the fact that the Internet was first imagined and framed as a ‘commons’ when actually it has created a stockpile for centralized control over (or the extraction and exploitation of) recursive, iterative and creative work. As with most computer technologies, the ‘freedom’ and ‘flexibility’ that these tools promise also depends on a loss of agency, control and freedom for many, in this case the artists, writers and researchers who have made their work accessible in this way. Thus, rather than fixate on the latest promissory technology or focus on a relatively small set of elite academic pursuits born out of a marriage between logic, statistics and modern digital computing, we explore AI as a diffuse set of technologies and systems of epistemic and political power that participate in broader historical trajectories than are traditionally offered, expanding the scope of what ‘history of AI’ is a history of .
This story, which begins from the widely deplored look of the graphing in Microsoft’s spreadsheet program, Excel, and extends back some seventy-five years, is about code that enables users to create visualizations without enough craft—at least in the eyes of their critics. It investigates two facets of data visualization since World War II: iterative analysis through graphical means and the making of business charts concerning numerical data. These two efforts automate some human skills. Both are seen as aberrant, dangerous, and in bad taste when they become too automatic, as users fail to reflect upon defaults. Both activities challenge the binary division between a “nonalgorithmic” culture of human judgments and a contemporary world subjected to hard, unaccountable logics. Telling a story of everyday cultures deemed to have gone bad, this article offers a history of envisioned users shaped through tools that could amplify virtues—and also vices—in thinking, depicting, and acting.
Abstract This chapter concerns the genesis and development of one of the foremost kinds of algorithms for supervised learning: decision trees. A series of researchers came obliquely to trees in the 1970s: a data-driven statistician, a machine learning expect focused on large data sets, social scientists unhappy with multivariate statistics, and a physicist interested mostly in computers who eventually was tenured in a statistics department. The history of trees is iterative: the implementation of algorithms on actually existing computers with various limitations drives the development and transformation of the techniques. Before the recent renaissance and current triumph of neural networks, decision trees were central to the transformation of artificial intelligence and machine learning of recent years: the shift in the central goal to a focus on prediction at the expense of concerns with human intelligibility, and a shift from symbolic interpretation to potent but inscrutable black boxes. Trees exploded in the late 1980s and 1990s as paragons of interpretable algorithms but developed in the late 1990s into an example of powerful but opaque ensemble models, predictive but almost unknowable. In some cases, techniques imported from academic statistics have made them even less attractive objects within traditional statistics, even as they are embedding in increasingly ubiquitous systems making judgments on our behalf. We need to explain, rather than take as given, the shift in values to prediction—to an instrumentalism—central to the ethos and practice of the contemporary data sciences. Studying trees, as this chapter does, will help us do this.
Journal of Ultrasound in Medicine · 2022-04-28 · 6 citations
article
We present a 10-patient case series supporting the use of the erector spinae plane block (ESPB) as a novel approach for the treatment of acute pain from renal colic. An in-plane needle approach was used with either transverse or longitudinal orientation of the ultrasound probe on the affected side, in either seated or prone patient position. These cases showed significant improvement in patient reported pain; suggesting that the ESPB can be used safely and effectively for either primary or adjunctive treatment of acute pain due to renal colic in the emergency department.
