Search

Your search keyword '"Cafagna, M."' showing total 15 results
Start Over Author "Cafagna, M."
15 results on '"Cafagna, M."'

1. Interpreting Vision and Language Generative Models with Semantic Visual Priors

Author

Cafagna, M, Rojas-Barahona, LM, van Deemter, K, Gatt, A, Cafagna, M, Rojas-Barahona, LM, van Deemter, K, and Gatt, A

Abstract

When applied to Image-to-text models, explainability methods have two challenges. First, they often provide token-by-token explanations namely, they compute a visual explanation for each token of the generated sequence. This makes explanations expensive to compute and unable to comprehensively explain the model's output. Second, for models with visual inputs, explainability methods such as SHAP typically consider superpixels as features. Since superpixels do not correspond to semantically meaningful regions of an image, this makes explanations harder to interpret. We develop a framework based on SHAP, that allows for generating comprehensive, meaningful explanations leveraging the meaning representation of the output sequence as a whole. Moreover, by exploiting semantic priors in the visual backbone, we extract an arbitrary number of features that allows the efficient computation of Shapley values on large-scale models, generating at the same time highly meaningful visual explanations. We demonstrate that our method generates semantically more expressive explanations than traditional methods at a lower compute cost and that it can be generalized to a large family of vision-language models.

Published
2023

2. HL Dataset: Grounding High-Level Linguistic Concepts in Vision

Author

Cafagna, M, van Deemter, K, Gatt, A, Cafagna, M, van Deemter, K, and Gatt, A

Abstract

Current captioning datasets focus on object-centric captions, describing the visible objects in the image, often ending up stating the obvious (for humans), e.g. “people eating food in a park”. Although these datasets are useful to evaluate the ability of Vision & Language models to recognize and describe visual content, they do not support controlled experiments involving model testing or fine-tuning, with more high-level captions, which humans find easy and natural to produce. For example, people often describe images based on the type of scene they depict (“people at a holiday resort”) and the actions they perform (“people having a picnic”). Such concepts are based on personal experience and contribute to forming common sense assumptions. We present the High-Level Dataset, a dataset extending 14997 images from the COCO dataset, aligned with a new set of 134,973 human-annotated (high-level) captions collected along three axes: scenes, actions and rationales. We further extend this dataset with confidence scores collected from an independent set of readers, as well as a set of narrative captions generated synthetically, by combining each of the three axes. We describe this dataset and analyse it extensively. We also present baseline results for the High-Level Captioning task.

Published
2023

3. HL Dataset: Grounding High-Level Linguistic Concepts in Vision

Author

Sub Natural Language Processing, Dep Informatica, Natural Language Processing, Cafagna, M, van Deemter, K, Gatt, A, Sub Natural Language Processing, Dep Informatica, Natural Language Processing, Cafagna, M, van Deemter, K, and Gatt, A

Published
2023

5. VALSE: A Task-independent benchmark for Vision and Language models centered on linguistic phenomena

Author

Parcalabescu, L, Cafagna, M, Muradjan, L, Frank, A, Calixto, I, Gatt, A, Parcalabescu, L, Cafagna, M, Muradjan, L, Frank, A, Calixto, I, and Gatt, A

Abstract

We propose VALSE (Vision And Language Structured Evaluation), a novel benchmark designed for testing general-purpose pretrained vision and language (V&L) models for their visio-linguistic grounding capabilities on specific linguistic phenomena. VALSE offers a suite of six tests covering various linguistic constructs. Solving these requires models to ground linguistic phenomena in the visual modality, allowing more fine-grained evaluations than hitherto possible. We build VALSE using methods that support the construction of valid foils, and report results from evaluating five widely-used V&L models. Our experiments suggest that current models have considerable difficulty addressing most phenomena. Hence, we expect VALSE to serve as an important benchmark to measure future progress of pretrained V&L models from a linguistic perspective, complementing the canonical task-centred V&L evaluations.

Published
2022

7. GePpeTto Carves Italian into a Language Model

Author

Mattei, L., Cafagna, M., FELICE DELLORLETTA, Nissim, M., and Guerini, M.

Subjects
FOS: Computer and information sciences, Twitter during Pandemic, Computer Science - Computation and Language, Automatic Sarcasm Detection, Linguistic Ostracism in Social Networks, AriEmozione, COVID-19, Linguistics, LAN000000, Quantitative Linguistic Investigations, Fine-grained sentiment analysis, Online Hate Speech, Computational Linguistics, DistilBERT, Depression from Social Media, Distributional Semantics, Gender Bias, CBX, AEREST, E3C Project, Multilingual NLU, TrAVaSI, Computation and Language (cs.CL)
Abstract

In the last few years, pre-trained neural architectures have provided impressive improvements across several NLP tasks. Still, generative language models are available mainly for English. We develop GePpeTto, the first generative language model for Italian, built using the GPT-2 architecture. We provide a thorough analysis of GePpeTto's quality by means of both an automatic and a human-based evaluation. The automatic assessment consists in (i) calculating perplexity across different genres and (ii) a profiling analysis over GePpeTto's writing characteristics. We find that GePpeTto's production is a sort of bonsai version of human production, with shorter but yet complex sentences. Human evaluation is performed over a sentence completion task, where GePpeTto's output is judged as natural more often than not, and much closer to the original human texts than to a simpler language model which we take as baseline.

Published
2021

8. VALSE: A Task-Independent Benchmark for Vision and Language Models Centered on Linguistic Phenomena

Author

Parcalabescu, L, Cafagna, M, Muradjan, L, Frank, A, Calixto, I, Gatt, A, Afd Intelligent Software Systems, Sub Natural Language Processing, Natural Language Processing, Afd Intelligent Software Systems, Sub Natural Language Processing, and Natural Language Processing

Subjects
FOS: Computer and information sciences, I.2.10, Computer Science - Computation and Language, I.2.7, Computer Vision and Pattern Recognition (cs.CV), Computer Science - Computer Vision and Pattern Recognition, 68Txx, Computation and Language (cs.CL)
Abstract

We propose VALSE (Vision And Language Structured Evaluation), a novel benchmark designed for testing general-purpose pretrained vision and language (V&L) models for their visio-linguistic grounding capabilities on specific linguistic phenomena. VALSE offers a suite of six tests covering various linguistic constructs. Solving these requires models to ground linguistic phenomena in the visual modality, allowing more fine-grained evaluations than hitherto possible. We build VALSE using methods that support the construction of valid foils, and report results from evaluating five widely-used V&L models. Our experiments suggest that current models have considerable difficulty addressing most phenomena. Hence, we expect VALSE to serve as an important benchmark to measure future progress of pretrained V&L models from a linguistic perspective, complementing the canonical task-centred V&L evaluations., Paper accepted for publication at ACL 2022 Main; 28 pages, 4 figures, 11 tables

Published
2021
Full Text
View/download PDF

9. Invisible to People but not to Machines: Evaluation of Style-aware Headline Generation in Absence of Reliable Human Judgment

Author

Mattei, L., Cafagna, M., FELICE DELLORLETTA, and Nissim, M.

Abstract

We automatically generate headlines that are expected to comply with the specific styles of two different Italian newspapers. Through a data alignment strategy and different training/testing settings, we aim at decoupling content from style and preserve the latter in generation. In order to evaluate the generated headlines’ quality in terms of their specific newspaper-compliance, we devise a fine-grained evaluation strategy based on automatic classification. We observe that our models do indeed learn newspaper-specific style.Importantly, we also observe that humans aren’t reliable judges for this task, since although familiar with the newspapers, they are notable to discern their specific styles even in the original human-written headlines. The utility of automatic evaluation goes therefore beyond saving the costs and hurdles of manual annotation, and deserves particular care in its design.

Published
2020

10. First Results from the TOTEM Experiment

Author

Antchev, G., Aspell, P., Atanassov, I., Avati, V., Berardi, V., Berretti, M., Bozzo, Marco, Buzzo, A., Calicchio, F. Cafagna M., Catanesi, M. G., Ciocci, M. A., Ferro, Fabrizio, Giani, S., Greco, V., Lami, S., Latino, G., LO VETERE, Maurizio, Macri, M., Magazzu, G., Minutoli, S., Notarnicola, G., Oliveri, E., Oriunno, M., Palazzi, P., Pedreschi, E., Quinto, M., Radicioni, E., Ravotti, F., Robutti, Enrico, Ruggiero, G., Sanguinetti, G., Santroni, Alberto, Scribano, A., Sette, Giuseppe, Spinella, F., and Turini, N.

Published
2011

11. TOTEM early measurements

Author

Antchev, G., Aspell, P., Atanassov, I., Avati, V., Berardi, V., Berretti, M., Bozzo, Marco, Buzzo, A., Calicchio, F. Cafagna M., Catanesi, M. G., Ciocci, M. A., Ferro, Fabrizio, Giani, S., Greco, V., Lami, S., Latino, G., LO VETERE, Maurizio, Macri, M., Magazzu, G., Minutoli, S., Notarnicola, G., Oliveri, E., Oriunno, M., Palazzi, P., Pedreschi, E., Quinto, M., Radicioni, E., Ravotti, F., Robutti, Enrico, Ruggiero, G., Sanguinetti, G., Santroni, Alberto, Scribano, A., Sette, Giuseppe, Spinella, F., and Turini, N.

Published
2009

12. Nuclearite search with the MACRO detector at Gran Sasso

Author

M. Ambrosio, R. Antolini, C. Aramo, G. Auriemma, A. Baldini, G. C. Barbarino, B. C. Barish, G. Battistoni, R. Bellotti, C. Bemporad, H. Bilokon, V. Bisi, C. Bloise, C. Bower, S. Bussino, F. Cafagna, M. Calicchio, D. Campana, M. Carboni, M. Castellano, S. Cecchini, F. Cei, V. Chiarella, B. C. Choudhary, S. Coutu, L. D. Benedictis, G. D. Cataldo, H. Dekhissi, E. D. Marzo, J. Derkaoui, M. D. Vincenzi, A. D. Credico, O. Erriquez, C. Favuzzi, C. Forti, P. Fusco, G. Giacomelli, G. Giannini, N. Giglietto, M. Giorgini, M. Grassi, L. Gray, A. Grillo, F. Guarino, P. Guarnaccia, C. Gustavino, A. Habig, K. Hanson, R. Heinz, Y. Huang, E. Iarocci, E. Katsavounidis, E. Kearns, H. Kim, S. Kyriazopoulou, E. Lamanna, C. Lane, T. Lari, D. S. Levin, P. Lipari, N. P. Longley, M. J. Longo, F. Maaroufi, G. Mandrioli, S. Manzoor, A. Margiotta, A. Marini, A. Marzari Chiesa, M. N. Mazziotta, D. G. Michael, S. Mikheyev, L. Miller, P. Monacelli, T. Montaruli, M. Monteno, S. Mufson, J. Musser, D. Nicolo, C. Orth, G. Osteria, M. Ouchrif, O. Palamara, V. Patera, L. Patrizii, R. Pazzi, C. W. Peck, S. Petrera, P. Pistilli, V. Popa, V. Pugliese, A. Raino, A. Rastelli, J. Reynoldson, F. Ronga, U. Rubizzo, C. Satriano, L. Satta, E. Scapparone, K. Scholberg, A. Sciubba, P. Serra, M. Severi, M. Sioli, M. Sitta, P. Spinelli, M. Spinetti, M. Spurio, R. Steinberg, J. L. Stone, L. R. Sulak, A. Surdo, G. Tarle, V. Togo, D. Ugolotti, M. Vakili, C. W. Walter, R. Webb, BERNARDINI, Paolo, DE MITRI, Ivan, MANCARELLA, Giovanni, MARTELLO, Daniele, PERRONE, Lorenzo, M., Ambrosio, R., Antolini, C., Aramo, G., Auriemma, A., Baldini, G. C., Barbarino, B. C., Barish, G., Battistoni, R., Bellotti, C., Bemporad, Bernardini, Paolo, H., Bilokon, V., Bisi, C., Bloise, C., Bower, S., Bussino, F., Cafagna, M., Calicchio, D., Campana, M., Carboni, M., Castellano, S., Cecchini, F., Cei, V., Chiarella, B. C., Choudhary, S., Coutu, L. D., Benedicti, G. D., Cataldo, H., Dekhissi, E. D., Marzo, DE MITRI, Ivan, J., Derkaoui, M. D., Vincenzi, A. D., Credico, O., Erriquez, C., Favuzzi, C., Forti, P., Fusco, G., Giacomelli, G., Giannini, N., Giglietto, M., Giorgini, M., Grassi, L., Gray, A., Grillo, F., Guarino, P., Guarnaccia, C., Gustavino, A., Habig, K., Hanson, R., Heinz, Y., Huang, E., Iarocci, E., Katsavounidi, E., Kearn, H., Kim, S., Kyriazopoulou, E., Lamanna, C., Lane, T., Lari, D. S., Levin, P., Lipari, N. P., Longley, M. J., Longo, F., Maaroufi, Mancarella, Giovanni, G., Mandrioli, S., Manzoor, A., Margiotta, A., Marini, Martello, Daniele, A., Marzari Chiesa, M. N., Mazziotta, D. G., Michael, S., Mikheyev, L., Miller, P., Monacelli, T., Montaruli, M., Monteno, S., Mufson, J., Musser, D., Nicolo, C., Orth, G., Osteria, M., Ouchrif, O., Palamara, V., Patera, L., Patrizii, R., Pazzi, C. W., Peck, Perrone, Lorenzo, S., Petrera, P., Pistilli, V., Popa, V., Pugliese, A., Raino, A., Rastelli, J., Reynoldson, F., Ronga, U., Rubizzo, C., Satriano, L., Satta, E., Scapparone, K., Scholberg, A., Sciubba, P., Serra, M., Severi, M., Sioli, M., Sitta, P., Spinelli, M., Spinetti, M., Spurio, R., Steinberg, J. L., Stone, L. R., Sulak, A., Surdo, G., Tarle, V., Togo, D., Ugolotti, M., Vakili, C. W., Walter, R., Webb, Ambrosio, M, Antolini, R, Aramo, C, Auriemma, G, Baldini, A, Barbarino, Gc, Barish, Bc, Battistoni, G, Bellotti, R, Bemporad, C, Bernardini, P, Bilokon, H, Bisi, V, Bloise, C, Bower, C, Bussino, Severino Angelo Maria, Cafagna, F, Calicchio, M, Campana, D, Carboni, M, Castellano, M, Cecchini, S, Cei, F, Chiarella, V, Choudhary, Bc, Coutu, S, De Benedictis, L, De Cataldo, G, Dekhissi, H, De Marzo, E, De Mitri, I, Derkaoui, J, De Vincenzi, M, Di Credico, A, Erriquez, O, Favuzzi, C, Forti, C, Fusco, P, Giacomelli, G, Giannini, G, Giglietto, N, Giorgini, M, Grassi, M, Gray, L, Grillo, A, Guarino, F, Guarnaccia, P, Gustavino, C, Habig, A, Hanson, K, Heinz, R, Huang, Y, Iarocci, E, Katsavounidis, E, Kearns, E, Kim, H, Kyriazopoulou, S, Lamanna, E, Lane, C, Lari, T, Levin, D, Lipari, P, Longley, Np, Longo, Mj, Maaroufi, F, Mancarella, G, Mandrioli, G, Manzoor, S, Margiotta, A, Marini, A, Martello, D, Marzari Chiesa, A, Mazziotta, Mn, Michael, Dg, Mikheyev, S, Miller, L, Monacelli, P, Montaruli, T, Monteno, M, Mufson, S, Musser, J, Nicolo, D, Orth, C, Osteria, G, Ouchrif, M, Palamara, O, Patera, V, Patrizii, L, Pazzi, R, Peck, Cw, Perrone, L, Petrera, S, Pistilli, P, Popa, V, Pugliese, V, Raino, A, Rastelli, A, Reynoldson, J, Ronga, F, Rubizzo, U, Satriano, C, Satta, L, Scapparone, E, Scholberg, K, Sciubba, A, Serra, P, Severi, M, Sioli, M, Sitta, M, Spinelli, P, Spinetti, M, Spurio, M, Steinberg, R, Stone, Jl, Sulak, Lr, Surdo, A, Tarle, G, Togo, V, Ugolotti, D, Vakili, M, Walter, Cw, Webb, R., and M. Ambrosio, R. Antolini, C. Aramo, G. Auriemma, A. Baldini, G. C. Barbarino, B. C. Barish, G. Battistoni, R. Bellotti, C. Bemporad, P. Bernardini, H. Bilokon, V. Bisi, C. Bloise, C. Bower, S. Bussino, F. Cafagna, M. Calicchio, D. Campana, M. Carboni, M. Castellano, S. Cecchini, F. Cei, V. Chiarella, B. C. Choudhary, S. Coutu, L. De Benedictis, G. De Cataldo, H. Dekhissi, C. De Marzo, I. De Mitri, J. Derkaoui, M. De Vincenzi, A. Di Credico, O. Erriquez, C. Favuzzi, C. Forti, P. Fusco, G. Giacomelli, G. Giannini , N. Giglietto, M. Giorgini, M. Grassi, L. Gray, A. Grillo, F. Guarino, P. Guarnaccia, C. Gustavino, A. Habig, K. Hanson, R. Heinz, Y. Huang, E. Iarocci, E. Katsavounidis, E. Kearns, H. Kim, S. Kyriazopoulou, E. Lamanna, C. Lane, T. Lari, D. S. Levin, P. Lipari, N. P. Longley, M. J. Longo, F. Maarou, G. Mancarella, G. Mandrioli, S. Manzoor, A. Margiotta, A. Marini, D. Martello, A. Marzari-Chiesa, M. N. Mazziotta, D. G. Michael, S. Mikheyev, L. Miller, P. Monacelli, T. Montaruli, M. Monteno, S. Mufson, J. Musser, D. Nicolo, C. Orth, G. Osteria, M. Ouchrif, O. Palamara, V. Patera, L. Patrizii, R. Pazzi, C. W. Peck, L. Perrone, S. Petrera, P. Pistilli, V. Popa, V. Pugliese, A. Raino, A. Rastelli, J. Reynoldson, F. Ronga, U. Rubizzo, C. Satriano, L. Satta, E. Scapparone, K. Scholberg, A. Sciubba, P. Serra, M. Severi, M. Sioli, M. Sitta, P. Spinelli, M. Spinetti, M. Spurio, R. Steinberg, J. L. Stone, L. R. Sulak, A. Surdo, G. Tarlè, V. Togo, D. Ugolotti, M. Vakili, C. W. Walter, R. Webb

Subjects
Physics and Astronomy (miscellaneous), MACRO, Engineering (miscellaneous), Gran Sasso
Abstract

In this paper we present the results of a search for nuclearites in the penetrating cosmic radiation using the scintillator and track-etch subdetectors of the MACRO apparatus. The analyses cover the beta = v/c range at the detector depth (3700 hg/cm(2)) 10(-5) < beta < 1; for beta = 2 x 10(-3) the flux limit is 2.7 x 10(-16) cm(-2)s(-1)sr(-1) for an isotropic flux of nuclearites, and twice this value for a flux of downgoing nuclearites.

Published
2000

13. Interpreting vision and language generative models with semantic visual priors.

Author

Cafagna M, Rojas-Barahona LM, van Deemter K, and Gatt A

Abstract

When applied to Image-to-text models, explainability methods have two challenges. First, they often provide token-by-token explanations namely, they compute a visual explanation for each token of the generated sequence. This makes explanations expensive to compute and unable to comprehensively explain the model's output. Second, for models with visual inputs, explainability methods such as SHAP typically consider superpixels as features. Since superpixels do not correspond to semantically meaningful regions of an image, this makes explanations harder to interpret. We develop a framework based on SHAP, that allows for generating comprehensive, meaningful explanations leveraging the meaning representation of the output sequence as a whole. Moreover, by exploiting semantic priors in the visual backbone, we extract an arbitrary number of features that allows the efficient computation of Shapley values on large-scale models, generating at the same time highly meaningful visual explanations. We demonstrate that our method generates semantically more expressive explanations than traditional methods at a lower compute cost and that it can be generalized to a large family of vision-language models., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2023 Cafagna, Rojas-Barahona, van Deemter and Gatt.)

Published
2023
Full Text
View/download PDF

14. Functional fibrinogen (FLEV-TEG) versus the Clauss method in an obstetric population: a comparative study.

Author

Spasiano A, Matellon C, Orso D, Brussa A, Cafagna M, Marangone A, Dogareschi T, Bove T, Giacomello R, Fontana D, Vetrugno L, and Della Rocca G

Subjects
Adult, Blood Coagulation Disorders blood, Blood Coagulation Disorders diagnosis, Blood Coagulation Tests methods, Cesarean Section, Female, Humans, Prospective Studies, Blood Coagulation physiology, Fibrinogen metabolism, Labor, Obstetric blood, Pregnancy blood, Thrombelastography methods
Abstract

Background: Hemostasis is the dynamic equilibrium between coagulation and fibrinolysis. During pregnancy, the balance shifts toward a hypercoagulative state; however placental abruption and abnormal placentations may lead to rapidly evolving coagulopathy characterized by the increased activation of procoagulant pathways. These processes can result in hypofibrinogenemia, with fibrinogen levels dropping to 2 g/L or less and an associated increased risk of post-partum hemorrhage. The aim of the present study was to evaluate the concordance between two methods of functional fibrinogen measurement: the Thromboelastography (TEG) method (also known as FLEV) vs. the Clauss method. Three patient groups were considered: healthy volunteers; non-pathological pregnant patients; and pregnant patients who went on to develop postpartum hemorrhage., Methods: A prospective observational study. Inclusion criteria were: healthy volunteer women of childbearing age, non-pathological pregnant women at term, and pregnant hemorrhagic patients subjected to elective or urgent caesarean section (CS), with blood loss exceeding 1000 mL. Exclusion criteria were age < 18 years, a history of coagulopathy, and treatment with contraceptives, anticoagulants, or antiplatelet agents., Results: Bland-Altman plots showed a significant overestimation with the FLEV method in all three patient groups: bias was - 133.36 mg/dL for healthy volunteers (95% IC: - 257.84; - 8.88. Critical difference: 124.48); - 56.30 mg/dL for healthy pregnant patients (95% IC: - 225.53; 112.93. Critical difference: 169.23); and - 159.05 mg/dL for hemorrhagic pregnant patients (95% IC: - 333.24; 15.148. Critical difference: 174.19). Regression analyses detected a linear correlation between FLEV and Clauss for healthy volunteers, healthy pregnant patients, and hemorrhagic pregnant patients (R 2 0.27, p value = 0.002; R 2 0.31, p value = 0.001; R 2 0.35, p value = 0.001, respectively). ANOVA revealed a statistically significant difference in fibrinogen concentration between all three patients groups when assayed using the Clauss method (p value < 0.001 for all the comparisons), but no statistically significant difference between the two patients groups of pregnant women when using the FLEV method., Conclusions: The FLEV method does not provide a valid alternative to the Clauss method due to the problem of fibrinogen overestimation, and for this reason it should not be recommended for the evaluation of patients with an increased risk of hypofibrinogenemia.

Published
2019
Full Text
View/download PDF

15. Unreliability of the %VO2 reserve versus %heart rate reserve relationship for aerobic effort relative intensity assessment in chronic heart failure patients on or off beta-blocking therapy.

Author

Mezzani A, Corrà U, Giordano A, Cafagna M, Adriano EP, and Giannuzzi P

Subjects
Aged, Carbazoles therapeutic use, Carvedilol, Chronic Disease, Ergometry, Exercise Test, Female, Humans, Male, Middle Aged, Propanolamines therapeutic use, Reproducibility of Results, Ventricular Dysfunction, Left, Adrenergic beta-Antagonists therapeutic use, Cardiac Output, Low drug therapy, Cardiac Output, Low physiopathology, Cardiomyopathy, Dilated drug therapy, Cardiomyopathy, Dilated physiopathology, Heart Rate, Physical Exertion physiology, Pulmonary Gas Exchange
Abstract

Background: Percentage of heart rate reserve (%HRR) has been shown to be equivalent to percentage of VO2 reserve (%VO2R) in normal individuals, but it is not clear whether this is also true for chronic heart failure (CHF) patients., Design: This study aimed to evaluate the reliability of the %VO2R versus %HRR relationship for aerobic effort relative intensity assessment in CHF patients., Methods: We studied 388 patients with CHF, of whom 213 were on carvedilol (CHF on betab) and 175 were off the drug (CHF off betab), and 40 normal individuals (N). %VO2R and %HRR values were evaluated at matched steps during cardiopulmonary exercise testing. The %VO2R versus %HRR regression was determined for each group (CHF, CHF on betab, CHF off betab, N) as a whole and as the mean of individual regressions., Results: %VO2R strongly correlated with %HRR in both N and CHF, at both group and mean individual regressions, with slope and y-intercept values significantly lower and higher, respectively, in CHF than N. The 95% prediction interval of %VO2R for a given %HRR value was 24% points in N but rose to 41 in CHF, and the mean regression line coincided with identity line in N but not in CHF. These results were independent of beta-blockade., Conclusions: In CHF patients the %VO2R versus %HRR relationship is unreliable for assessment of aerobic effort relative intensity, because of a large prediction interval of %VO2R and lack of coincidence with the identity line, independently of beta-blocking therapy. This implies that the %VO2R versus %HRR relationship should be determined directly in each patient.

Published
2007
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results