Browsing by Author "Besa Correa, Cecilia"

Now showing 1 - 12 of 12
  • Thumbnail Image
    Item
    Detection of Liver Fibrosis Using Qualitative and Quantitative MR Elastography Compared to Liver Surface Nodularity Measurement, Gadoxetic Acid Uptake, and Serum Markers
    (2018) Besa Correa, Cecilia; Wagner, Mathilde; Lo, Grace; Gordic, Sonja; Chatterji, Manjil; Kennedy, Paul; Stueck, Ashley; Thung, Swan; Babb, James; Smith, Andrew; Taouli, Bachir
  • Thumbnail Image
    Item
    Fecaloma gigante en un paciente adolescente
    (2014) Caselli, G.; Besa Correa, Cecilia; Miguieles Cocco, Rodrigo Alfonso
  • Thumbnail Image
    Item
    Hemangioma cavernoso rectal. Caso clínico
    (2010) Urrejola Schmied, Gonzalo Ignacio; Caselli, G.; Besa Correa, Cecilia; Torrealba, I.; Zuniga, A. M.
  • Thumbnail Image
    Item
    Hepatocellular carcinoma detection in liver cirrhosis: diagnostic performance of contrast-enhanced CT vs. MRI with extracellular contrast vs. gadoxetic acid
    (2020) Semaan, S; Violi, NV; Lewis, S; Chatterji, M; Song, CT; Besa Correa, Cecilia; Babb, JS; Fiel, MI; Schwartz, M; Thung, S; Sirlin, CB; Taouli, B
  • Thumbnail Image
    Item
    MRI radiomics features predict immuno-oncological characteristics of hepatocellular carcinoma
    (2020) Hectors, S. J.; Lewis, S.; Besa Correa, Cecilia; King, M. J.; Said, D.; Putra, J.; Ward, S.; Higashi, T.; Thung, S.; Yao, S.; Laface, I.; Schwartz, M.; Gnjatic, S.; Merad, M.; Hoshida, Y.; Taouli, B.
  • Thumbnail Image
    Item
    Multiparametric FDG-PET/MRI of Hepatocellular Carcinoma : Initial Experience
    (2018) Hectors, Stefanie J.; Wagner, Mathilde; Besa Correa, Cecilia; Huang, Wei; Taouli, Bachir
  • Thumbnail Image
    Item
    Noninvasive prediction of portal pressure with MR elastography and DCE-MRI of the liver and spleen : Preliminary results
    (2018) Wagner, Mathilde; Hectors, Stefanie; Bane, Octavia; Gordic, Sonja; Kennedy, Paul; Besa Correa, Cecilia; Schiano, Thomas D.; Thung, Swan; Fischman, Aaron; Taouli, Bachir
  • Thumbnail Image
    Item
    Prediction of hepatocellular carcinoma response to 90Yttrium radioembolization using volumetric ADC histogram quantification: preliminary results
    (2019) Gordic, Sonja.; Wagner, Mathilde.; Zanato, Riccardo.; Hectors, Stefanie.; Besa Correa, Cecilia; Kihira, Shingo.; Kim, Edward.; Taouli, Bachir.
    Abstract Purpose To assess the predictive value of volumetric apparent diffusion coefficient (vADC) histogram quantification obtained before and 6 weeks (6w) post-treatment for assessment of hepatocellular carcinoma (HCC) response to 90Yttrium radioembolization (RE). Methods In this retrospective study, 22 patients (M/F 15/7, mean age 65y) who underwent lobar RE were included between October 2013 and November 2014. All patients underwent routine liver MRI pre-treatment and 6w after RE. Two readers assessed index tumor response at 6 months after RE in consensus, using mRECIST criteria. vADC histogram parameters of index tumors at baseline and 6w, and changes in vADC (ΔvADC) histogram parameters were calculated. The predictive value of ADC metrics was assessed by logistic regression with stepwise parameter selection and ROC analyses. Results Twenty two HCC lesions (mean size 3.9 ± 2.9 cm, range 1.2–12.3 cm) were assessed. Response at 6 months was as follows: complete response (CR, n = 6), partial response (PR, n = 3), stable disease (SD, n = 12) and progression (PD, n = 1). vADC median/mode at 6w (1.81–1.82 vs. 1.29–1.35 × 10− 3 mm2/s) and ΔvADC median/max (27–44% vs. 0–10%) were significantly higher in CR/PR vs. SD/PD (p = 0.011–0.036), while there was no significant difference at baseline. Logistic regression identified vADC median at 6w as an independent predictor of response (CR/PR) with odds ratio (OR) of 3.304 (95% CI: 1.099–9.928, p = 0.033) and AUC of 0.77. ΔvADC mean was identified as an independent predictor of CR with OR of 4.153 (95%CI: 1.229–14.031, p = 0.022) and AUC of 0.91. Conclusion Diffusion histogram parameters obtained at 6w and early changes in ADC from baseline are predictive of subsequent response of HCCs treated with RE, while pre-treatment vADC histogram parameters are not. These results need confirmation in a larger study. Trial registration This retrospective study was IRB-approved and the requirement for informed consent was waived.
  • Thumbnail Image
    Item
    Prediction of the histopathologic findings of intrahepatic cholangiocarcinoma : qualitative and quantitative assessment of diffusion-weighted imaging
    (2018) Lewis, Sara; Besa Correa, Cecilia; Wagner, Mathilde; Jhaveri, Kartik; Kihira, Shingo; Zhu, Hongfa; Sadoughi, Nima; Fischer, Sandra; Srivastava, Amogh; Yee, Eric; Mortele, Koenraad; Babb, James; Thung, Swan; Taouli, Bachir
  • Thumbnail Image
    Item
    Preliminary study of ventilation with 4 ml/kg tidal volume in acute respiratory distress syndrome: feasibility and effects on cyclic recruitment - derecruitment and hyperinflation
    (2013) Retamal Montes, Jaime; Libuy, Javiera; Jiménez, Magdalena; Delgado, Matías; Besa Correa, Cecilia; Bugedo Tarraza, Guillermo; Bruhn, Alejandro
    Abstract Introduction Cyclic recruitment-derecruitment and overdistension contribute to ventilator-induced lung injury. Tidal volume (Vt) may influence both, cyclic recruitment-derecruitment and overdistension. The goal of this study was to determine if decreasing Vt from 6 to 4 ml/kg reduces cyclic recruitment-derecruitment and hyperinflation, and if it is possible to avoid severe hypercapnia. Methods Patients with pulmonary acute respiratory distress syndrome (ARDS) were included in a crossover study with two Vt levels: 6 and 4 ml/kg. The protocol had two parts: one bedside and other at the CT room. To avoid severe hypercapnia in the 4 ml/kg arm, we replaced the heat and moisture exchange filter by a heated humidifier, and respiratory rate was increased to keep minute ventilation constant. Data on lung mechanics and gas exchange were taken at baseline and after 30 minutes at each Vt (bedside). Thereafter, a dynamic CT (4 images/sec for 8 sec) was taken at each Vt at a fixed transverse region between the middle and lower third of the lungs. Afterward, CT images were analyzed and cyclic recruitment-derecruitment was determined as non-aerated tissue variation between inspiration and expiration, and hyperinflation as maximal hyperinflated tissue at end-inspiration, expressed as % of lung tissue weight. Results We analyzed 10 patients. Decreasing Vt from 6 to 4 ml/kg consistently decreased cyclic recruitment-derecruitment from 3.6 (2.5 to 5.7) % to 2.9 (0.9 to 4.7) % (P <0.01) and end-inspiratory hyperinflation from 0.7 (0.3 to 2.2) to 0.6 (0.2 to 1.7) % (P = 0.01). No patient developed severe respiratory acidosis or severe hypercapnia when decreasing Vt to 4 ml/kg (pH 7.29 (7.21 to 7.46); PaCO2 48 (26 to 51) mmHg). Conclusions Decreasing Vt from 6 to 4 ml/kg reduces cyclic recruitment-derecruitment and hyperinflation. Severe respiratory acidosis may be effectively prevented by decreasing instrumental dead space and by increasing respiratory rate.Abstract Introduction Cyclic recruitment-derecruitment and overdistension contribute to ventilator-induced lung injury. Tidal volume (Vt) may influence both, cyclic recruitment-derecruitment and overdistension. The goal of this study was to determine if decreasing Vt from 6 to 4 ml/kg reduces cyclic recruitment-derecruitment and hyperinflation, and if it is possible to avoid severe hypercapnia. Methods Patients with pulmonary acute respiratory distress syndrome (ARDS) were included in a crossover study with two Vt levels: 6 and 4 ml/kg. The protocol had two parts: one bedside and other at the CT room. To avoid severe hypercapnia in the 4 ml/kg arm, we replaced the heat and moisture exchange filter by a heated humidifier, and respiratory rate was increased to keep minute ventilation constant. Data on lung mechanics and gas exchange were taken at baseline and after 30 minutes at each Vt (bedside). Thereafter, a dynamic CT (4 images/sec for 8 sec) was taken at each Vt at a fixed transverse region between the middle and lower third of the lungs. Afterward, CT images were analyzed and cyclic recruitment-derecruitment was determined as non-aerated tissue variation between inspiration and expiration, and hyperinflation as maximal hyperinflated tissue at end-inspiration, expressed as % of lung tissue weight. Results We analyzed 10 patients. Decreasing Vt from 6 to 4 ml/kg consistently decreased cyclic recruitment-derecruitment from 3.6 (2.5 to 5.7) % to 2.9 (0.9 to 4.7) % (P <0.01) and end-inspiratory hyperinflation from 0.7 (0.3 to 2.2) to 0.6 (0.2 to 1.7) % (P = 0.01). No patient developed severe respiratory acidosis or severe hypercapnia when decreasing Vt to 4 ml/kg (pH 7.29 (7.21 to 7.46); PaCO2 48 (26 to 51) mmHg). Conclusions Decreasing Vt from 6 to 4 ml/kg reduces cyclic recruitment-derecruitment and hyperinflation. Severe respiratory acidosis may be effectively prevented by decreasing instrumental dead space and by increasing respiratory rate.Abstract Introduction Cyclic recruitment-derecruitment and overdistension contribute to ventilator-induced lung injury. Tidal volume (Vt) may influence both, cyclic recruitment-derecruitment and overdistension. The goal of this study was to determine if decreasing Vt from 6 to 4 ml/kg reduces cyclic recruitment-derecruitment and hyperinflation, and if it is possible to avoid severe hypercapnia. Methods Patients with pulmonary acute respiratory distress syndrome (ARDS) were included in a crossover study with two Vt levels: 6 and 4 ml/kg. The protocol had two parts: one bedside and other at the CT room. To avoid severe hypercapnia in the 4 ml/kg arm, we replaced the heat and moisture exchange filter by a heated humidifier, and respiratory rate was increased to keep minute ventilation constant. Data on lung mechanics and gas exchange were taken at baseline and after 30 minutes at each Vt (bedside). Thereafter, a dynamic CT (4 images/sec for 8 sec) was taken at each Vt at a fixed transverse region between the middle and lower third of the lungs. Afterward, CT images were analyzed and cyclic recruitment-derecruitment was determined as non-aerated tissue variation between inspiration and expiration, and hyperinflation as maximal hyperinflated tissue at end-inspiration, expressed as % of lung tissue weight. Results We analyzed 10 patients. Decreasing Vt from 6 to 4 ml/kg consistently decreased cyclic recruitment-derecruitment from 3.6 (2.5 to 5.7) % to 2.9 (0.9 to 4.7) % (P <0.01) and end-inspiratory hyperinflation from 0.7 (0.3 to 2.2) to 0.6 (0.2 to 1.7) % (P = 0.01). No patient developed severe respiratory acidosis or severe hypercapnia when decreasing Vt to 4 ml/kg (pH 7.29 (7.21 to 7.46); PaCO2 48 (26 to 51) mmHg). Conclusions Decreasing Vt from 6 to 4 ml/kg reduces cyclic recruitment-derecruitment and hyperinflation. Severe respiratory acidosis may be effectively prevented by decreasing instrumental dead space and by increasing respiratory rate.
  • Thumbnail Image
    Item
    Training and intrinsic evaluation of lightweight word embeddings for the clinical domain in Spanish
    (2022) Chiu, Carolina; Villena, Fabián; Martin, Kinan; Núñez, Fredy R.; Besa Correa, Cecilia; Dunstan, Jocelyn
    Resources for Natural Language Processing (NLP) are less numerous for languages different from English. In the clinical domain, where these resources are vital for obtaining new knowledge about human health and diseases, creating new resources for the Spanish language is imperative. One of the most common approaches in NLP is word embeddings, which are dense vector representations of a word, considering the word's context. This vector representation is usually the first step in various NLP tasks, such as text classification or information extraction. Therefore, in order to enrich Spanish language NLP tools, we built a Spanish clinical corpus from waiting list diagnostic suspicions, a biomedical corpus from medical journals, and term sequences sampled from the Unified Medical Language System (UMLS). These three corpora can be used to compute word embeddings models from scratch using Word2vec and fastText algorithms. Furthermore, to validate the quality of the calculated embeddings, we adapted several evaluation datasets in English, including some tests that have not been used in Spanish to the best of our knowledge. These translations were validated by two bilingual clinicians following an ad hoc validation standard for the translation. Even though contextualized word embeddings nowadays receive enormous attention, their calculation and deployment require specialized hardware and giant training corpora. Our static embeddings can be used in clinical applications with limited computational resources. The validation of the intrinsic test we present here can help groups working on static and contextualized word embeddings. We are releasing the training corpus and the embeddings within this publication.
  • Thumbnail Image
    Item