1,720,953 research outputs found
The Use of Artificial Intelligence for Diagnosis and Outcome Prediction in Primary Care
No full textIn recent years, the ability of Artificial Intelligence (AI) to analyze and predict various
variables within medicine has advanced significantly. Part of this progress can be
attributed to the availability of larger electronic datasets; however, there is still a
shortage of high-quality data for training AI models for use within the healthcare system.
Electronic health records contain vast amounts of data, and with increased
interoperability, diverse datasets are emerging that can be cross-referenced and
potentially used to train AI models. Doctors' summary notes from patient interviews are
a major component of medical records and include descriptions of the patient's
medical history, examination, findings, and the doctor’s assessment and plan. These
notes should, therefore, contain all the necessary information to train AI models to
predict patient diagnoses and outcomes.
The objective of this doctoral thesis was: 1) to investigate whether doctors’ summary
notes from patient medical records could be used to train AI models to predict
diagnoses and outcomes for patients in primary care; and 2) to study which factors the
models use to reach a conclusion and compare these to the elements doctors rely on.
The thesis is based on studies described in three papers—the first two were
retrospective, and the third was prospective. In the first study, an AI model was trained
to predict primary headache diagnoses based on labeled diagnostic features from
doctors' summary notes following patient interviews in primary care, which resulted in
one of four common primary headache diagnoses. The study compared the model’s
performance to that of three resident physicians and three family medicine specialists.
The model's internal functionality was also evaluated using Shapley Additive
Explanations (SHAP) values, which indicate which input features have the most
significant impact on each output. In the second paper, an AI model was trained on
Clinical Features (CFs) from labeled notes of doctors who treated primary care patients
diagnosed with one of the following respiratory symptom codes over a specific period:
J00, J15, J20, J44, and J45. In both the first and second studies, a methodology was
applied to remove low-content text notes and maximize the quality of the dataset from
which the models learned. The third study evaluated the model trained in the second
study, with minor adjustments, using a prospective approach in a single primary care
clinic in Iceland's capital area.
In the first study, the model achieved a higher weighted average sensitivity, positive
predictive value, and Matthews Correlation Coefficient (MCC) than the weighted
average of the doctors. The specificity of five out of six doctors was higher than that of
the model. SHAP value analysis indicated that the model relies on similar diagnostic
features as doctors for each diagnosis. In the second study, the model categorized
patients into risk groups, where the outcomes of patients in lower-risk groups reflected
those likely to have mild symptoms that resolve without intervention. No cases of
pneumonia or patients with lung infiltrates on Chest X-Rays (CXRs) were detected in the
lower-risk groups. In the third study, one patient with pneumonia was categorized in a
lower-risk group but was later found to have a normal lung CXR, suggesting a likely
misdiagnosis. All other patients with pneumonia on CXR were in high-risk groups. Two
patients were diagnosed with lung cancer, both in the highest risk group.
The results of the first study indicated that the AI model performed as well or slightly
better than the groups of GP trainees and GP specialists in diagnosing primary
headaches. SHAP value analysis showed that the models rely on similar clinical features
to doctors when making diagnoses. The results of the second study suggested that the
AI model could safely triage primary care patients with respiratory infections. The
model categorized patients with truly severe respiratory conditions into high-risk
groups, while those with milder symptoms were placed in low-risk groups. The third
study demonstrated that the model could stratify patients in real clinical settings so that
patients with severe respiratory conditions were categorized as high risk, while those
with mild symptoms were categorized as low risk. Two patients diagnosed with
pneumonia by doctors, whom the model classified as low risk, were subsequently
found to have normal lung images.
The doctoral thesis concludes that AI models trained on clinical features extracted from
physicians’ summary notes can have significant utility in primary healthcare.Á undanförnum árum hefur getu gervigreindar til þess að greina og spá fyrir um
útkomur sjúklinga verulega fleygt fram. Hluta af þessum framförum má þakka aðgengi
að stærri rafrænum gagnasöfnum en skortur er á stórum gæðamiklum gagnasöfnum til
þess að þjálfa gervigreindarlíkön á til notkunar innan heilbrigðiskerfisins. Rafrænar
sjúkraskrár búa yfir gríðarmiklum gögnum og með samtengingu kerfa geta myndast
fjölbreytt gagnasöfn sem má samkeyra og hugsanlega nýta til þjálfunar
gervigreindarlíkana. Samantektarnótur lækna úr viðtölum við sjúklinga eru stór hluti
sjúkraskrár í heilsugæslu og innihalda lýsingu á sögu sjúklings, skoðun, niðurstöðum og
ályktun og áætlun læknis. Slíkar nótur ættu því að innihalda allar upplýsingar sem þörf
er á til að þjálfa gervigreindarlíkön til að spá fyrir um greiningar og útkomur sjúklinga.
Markmið þessarar doktorsritgerðar var: 1) að rannsaka hvort hægt væri að nýta
textagögn úr sjúkraskrám sjúklinga til þess að þjálfa gervigreindaralíkön til þess að spá
fyrir um greiningar og horfur sjúklinga í heilsugæslu; 2) að skoða hvaða breytur hafa
áhrif á úttak líkananna og bera saman við breytur sem læknar styðjast við; og 3) kanna
frammistöðu líkananna við greiningar og forspár fyrir horfur sjúklinga í heilsugæslu.
Ritgerðin byggir á rannsóknum sem hefur verið lýst í þremur greinum—tvær fyrstu
rannsóknirnar eru aftursýnar en sú þriðja er framsýn. Í fyrstu greininni var
gervigreindarlíkan þjálfað til þess að spá fyrir um frumkomna höfuðverkjagreiningu út
frá merktum greiningarsérkennum úr samantektarnótu læknis eftir viðtöl við sjúklinga
sem fengu eina af fjórum algengustu greiningum slíkra höfuðverkja í heilsugæslu. Í
greininni var frammistaða líkansins borin saman við frammistöðu þriggja sérnámslækna
í heimilislækningum og þriggja sérfræðimenntaðra heimilislækna. Einnig var innri
virkni líkansins metin með SHapley Additive exPlanations (SHAP) gildum sem gáfu til
kynna hvaða þættir inntaksins höfðu mest áhrif á hvert úttak fyrir sig. Í annarri greininni
var gervigreindarlíkan þjálfað á greiningarsérkennum úr merktum nótum frá læknum
sem hittu sjúklinga í heilsugæslu sem fengu einn af eftirfarandi International
Classification of Diseases (ICD) kóðum á ákveðnu tímabili: J00, J15, J20, J44 og J45. Í
tveimur fyrstu greinunum var ákveðinni aðferðarfræði beitt til að fjarlægja innihaldslitlar
textanótur og hámarka gæði gagnasafnsins sem líkönin læra af. Í þriðju greininni var
frammistaða líkansins sem þjálfað var í rannsókn tvö metin, eftir smávægilega aðlögun,
með framsýnum hætti á einni heilsugæslustöð á höfuðborgarsvæðinu á Íslandi.
Í fyrstu rannsókninni náði líkanið hærra vegnu meðaltali í næmi, jákvæðu forspárgildi
og Matthews Correlation Coefficient (MCC) en vegið meðaltal læknanna. Sértæki fimm
lækna af sex var hærra en líkansins. Greining á SHAP gildum sýndi að líkanið styðst við
svipuð greiningarsérkenni og læknir gerir fyrir hverja greiningu. Í rannsókn tvö raðaði líkanið sjúklingum, með einkenni um öndunarfærasýkingu, í áhættuhópa þannig að
útkomur sjúklinga sem voru í lægri áhættuhópum endurspegluðu hóp sem líklega var
með væg einkenni sem hefðu gengið yfir án inngrips. Engir sjúklingar í lægri
áhættuhópum voru með lungnabólgu greinda af lækni eða á röntgenmynd. Í síðustu
rannsókninni var einn sjúklingur greindur með lungnabólgu í lægri áhættuhópum en
hann reyndist vera með eðlilega röntgemynd af lungum og því líklega rangt greindur.
Allir aðrir sjúklingar með lungnabólgu á röntgenmynd voru í há-áhættu hópum. Tveir
sjúklingar voru greindir með lungnakrabbamein, báðir í hæsta áhættuhóp.
Niðurstöður fyrstu rannsóknarinnar voru þær að gervigreindarlíkanið stóð sig jafn vel
eða ívið betur í greiningum á frumkomnum höfuðverkjum en hóparnir tveir af
sérnámslæknum og sérfræðingum í heimilislækningum. Greining á SHAP gildum leiddi
í ljós að líkönin studdust við mjög svipuð greiningarsérkenni og læknar við greiningar.
Niðurstöður annarrar rannsóknarinnar bentu til þess að gervigreindarlíkaninu tækist að
einkennastiga sjúklinga með öndunarfærasýkingar í heilsugæslu með öruggum hætti.
Líkanið raðaði sjúklingum sem reyndust raunverulega vera með alvarlega
öndunarfærasjúkdóma í há-áhættu hópa en sjúklingum með vægari einkenni í lág-
áhættu hópa. Þriðja rannsóknin sýndi fram á að líkaninu virtist takast að einkennastiga
sjúklinga í raunverulegum klínískum aðstæðum þannig að sjúklingar með alvarlegar
öndunarfærasýkingar voru flokkaðir í háa áhættu en sjúklingar með væg einkenni í lága
áhættu. Tveir sjúklingar sem læknar greindu með lungnabólgu en líkanið flokkaði í lág-
áhættu hóp reyndust vera með eðlilegar lungnamyndir.
Niðurstöður doktorsritgerðarinnar benda til þess að gervigreindarlíkön, sem þjálfuð eru
á greiningarsérkennum úr samantektarnótum lækna, geti haft verulegt notagildi í
heilsugæslu.Vísindasjóður Félags íslenskra heimilislækna, Ranní
Going Beyond Counting First Authors in Author Co-citation Analysis
Full text linkThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
Variations on the Author
Full text link“Variations on the Author” discusses two of Eduardo Coutinho’s recent films (Um Dia na Vida, from 2010, and Últimas Conversas, posthumously released in 2015) and their contribution to the general question of documentary authorship. The director’s filmography is characterized by a consistent yet self-effacing form of authorial self-inscription: Coutinho often features as an interviewer that rather than express opinions propels discourses; an interviewer that is good at listening. This mode of self-inscription characterizes him as an author who is not expressive but who is nonetheless markedly present on the screen. In Um Dia na Vida, however, Coutinho is completely absent form the image, while Últimas Conversas, on the contrary, includes a confessional prologue that moves the director from the margins to the center of his films. This article examines the ways in which these works stand out in the filmography of a director who offers new insights into the notion of cinematic authorship
Appropriate Similarity Measures for Author Cocitation Analysis
Full text linkWe provide a number of new insights into the methodological discussion about author cocitation analysis. We first argue that the use of the Pearson correlation for measuring the similarity between authors’ cocitation profiles is not very satisfactory. We then discuss what kind of similarity measures may be used as an alternative to the Pearson correlation. We consider three similarity measures in particular. One is the well-known cosine. The other two similarity measures have not been used before in the bibliometric literature. Finally, we show by means of an example that our findings have a high practical relevance.information science;Pearson correlation;cosine;similarity measure;author cocitation analysis
Dispelling the Myths Behind First-author Citation Counts
Full text linkWe conducted a full-scale evaluative citation analysis study of scholars in the XML research field to explore just how different from each other author rankings resulting from different citation counting methods actually are, and to demonstrate the capability of emerging data and tools on the Web in supporting more realistic citation counting methods. Our results contest some common arguments for the continued
use of first-author citation counts in the evaluation of scholars, such as high correlations between author rankings by first-author citation counts and other citation
counting methods, and high costs of using more realistic citation counting methods that are not well-supported by the ISI databases. It is argued that increasingly available digital full text research papers make it possible for citation analysis studies to go beyond what the ISI databases have directly supported and to employ more
sophisticated methods
koamabayili/VECTRON-author-checklist: VECTRON author checklist
No full textWe have done our best to complete the author checklist relating to the use of animals in the hut study. Note that the objective for the hut study was to evaluate the IRS treatment applications for residual efficacy against Anopheles mosquitoes, including the local An. coluzzii mosquito population. Cows were only used to attract mosquitoes into the huts and no tests were carried out directly on the cows. The author checklist is intended for use with studies where experiments are carried out on animals, which is why we have had such difficulty in completing this for the hut study, as many of the questions do not relate to how the cows were used
Author-wise bibliometric analysis based on entropy.
No full textAuthor-wise bibliometric analysis based on entropy.</p
Author Under Sail The Imagination of Jack London, 1893-1902
No full textIn Author Under Sail, Jay Williams offers the first complete literary biography of Jack London as a professional writer engaged in the labor of writing. It examines the authorial imagination in London's work, the use of imagination in both his fiction and nonfiction, and the ways he defined imagination in the creative process in his business dealings with his publishers, editors, and agents. In this first volume of a two-volume biography, Williams traverses the years 1893 to 1902, from London's "Story of a Typhoon" to The People of the Abyss. The Jack London who emerges in the pages of Author Under Sail is a writer whose partnership with publishers, most notably his productive alliance with George Brett of Macmillan, was one of the most formative in American literary history. London pioneered many author models during the heyday of realism and naturalism, blurring the boundaries of these popular genres by focusing on absorption and theatricality and the representation of the seen and unseen. London created an impassioned, sincere, and extremely personal realism unlike that of other American writers of the time. Author Under Sail is a literary tour de force that reveals the full range of London as writer, creative citizen, and entrepreneur at the same time it sheds light on the maverick side of machine-age literature.Intro -- Title Page -- Copyright Page -- Dedication -- Contents -- Acknowledgments -- Introduction -- 1. Spirit Truth -- 2. From Absorption to Theatricality and Back Again -- 3. "I Will Build a New Present" -- 4. Sons as Authors -- 5. Fathers as Publishers -- 6. The Daughter as Author -- 7. Lovers as Authors -- 8. At Sea with the Family -- 9. Yellow News, Yellow Stories -- 10. The Return Home -- Notes -- Bibliography -- Index -- About Jay WilliamsIn Author Under Sail, Jay Williams offers the first complete literary biography of Jack London as a professional writer engaged in the labor of writing. It examines the authorial imagination in London's work, the use of imagination in both his fiction and nonfiction, and the ways he defined imagination in the creative process in his business dealings with his publishers, editors, and agents. In this first volume of a two-volume biography, Williams traverses the years 1893 to 1902, from London's "Story of a Typhoon" to The People of the Abyss. The Jack London who emerges in the pages of Author Under Sail is a writer whose partnership with publishers, most notably his productive alliance with George Brett of Macmillan, was one of the most formative in American literary history. London pioneered many author models during the heyday of realism and naturalism, blurring the boundaries of these popular genres by focusing on absorption and theatricality and the representation of the seen and unseen. London created an impassioned, sincere, and extremely personal realism unlike that of other American writers of the time. Author Under Sail is a literary tour de force that reveals the full range of London as writer, creative citizen, and entrepreneur at the same time it sheds light on the maverick side of machine-age literature.Description based on publisher supplied metadata and other sources.Electronic reproduction. Ann Arbor, Michigan : ProQuest Ebook Central, YYYY. Available via World Wide Web. Access may be limited to ProQuest Ebook Central affiliated libraries
- …
