Roberto Franchini
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Roberto Franchini has a total of 52 co-authors. Here are the top co-authors by number of shared publications:
- Francesco Conversano
National Research Council – Institute of Clinical Physiology, Lecce, Italy
20 shared publications - Sergio Casciaro
Institute of Clinical Physiology, National Research Council, c/o Campus Ecotekne (Ed. A7), Via Prov.Le Lecce-Monteroni, 73100, Lecce, Italy.
15 shared publications Ernesto Casciaro
National Research Council Institute of Clinical Physiology Lecce Italy
10 shared publicationsAntonio Greco
Echolight Srl, Lecce, Italy
6 shared publicationsE. Casciaro
Institute of Clinical Physiology, National Council of Research, Lecce, Italy
5 shared publicationsGiulia Soloperto
National Research Council, Institute of Clinical Physiology, Lecce, Italy
5 shared publications
Latest Publications
- 30 November 2025
- Published by International Journal for Research in Applied Science and Engineering Technology in International Journal for Research in Applied Science and Engineering Technology
- Vol. 13 (11) , 2691-2697
Advancing Neurodiagnostic with Generative AI: Synthetic Data-Driven Lesion Detection in MRI
Recent advancements in generative artificial intelligence (AI) are redefining the landscape of non-invasive diagnostics in medical imaging, with significant impact on brain lesion detection via Magnetic Resonance Imaging (MRI). Conventional radiological assessment relies heavily on expert interpretation and manual lesion delineation—procedures that are inherently time-consuming and susceptible to inter- and intra-observer variability. In contrast, generative AI introduces automated, datadriven solutions capable of enhancing diagnostic precision through image synthesis, augmentation, and high-resolution segmentation. State-of-the-art architectures, including Generative Adversarial Networks (GANs) and diffusion-based models, can simulate anatomically consistent MRI scans, reconstruct obscured or missing structures, and amplify subtle pathological signatures that may evade traditional evaluation. By learning complex mappings between healthy and pathological tissue distributions, these models generate high-fidelity synthetic data that benefit both clinical prediction workflows and the training of conventional discriminative algorithms. Additionally, generative augmentation alleviates the scarcity of labeled datasets—a persistent limitation in medical imaging—by producing realistic and diverse lesion-focused samples. This study introduces a generative AI-driven framework for automated brain lesion recognition and classification, comprising standardized preprocessing, targeted data augmentation, and a hybrid discriminative–generative modeling pipeline. The system emphasizes robustness and clinical transparency through explainable inference modules and integrated uncertainty quantification. Experimental results demonstrate that generative learning markedly improves segmentation and classification accuracy, particularly in cases involving rare or morphologically ambiguous lesions. These findings support the integration of generative AI as a cornerstone technology for next-generation precise, scalable, and non-invasive neurodiagnostic workflows
- 30 November 2025
- Published by International Journal for Research in Applied Science and Engineering Technology in International Journal for Research in Applied Science and Engineering Technology
- Vol. 13 (11) , 2691-2697
Advancing Neurodiagnostic with Generative AI: Synthetic Data-Driven Lesion Detection in MRI
Recent advancements in generative artificial intelligence (AI) are redefining the landscape of non-invasive diagnostics in medical imaging, with significant impact on brain lesion detection via Magnetic Resonance Imaging (MRI). Conventional radiological assessment relies heavily on expert interpretation and manual lesion delineation—procedures that are inherently time-consuming and susceptible to inter- and intra-observer variability. In contrast, generative AI introduces automated, datadriven solutions capable of enhancing diagnostic precision through image synthesis, augmentation, and high-resolution segmentation. State-of-the-art architectures, including Generative Adversarial Networks (GANs) and diffusion-based models, can simulate anatomically consistent MRI scans, reconstruct obscured or missing structures, and amplify subtle pathological signatures that may evade traditional evaluation. By learning complex mappings between healthy and pathological tissue distributions, these models generate high-fidelity synthetic data that benefit both clinical prediction workflows and the training of conventional discriminative algorithms. Additionally, generative augmentation alleviates the scarcity of labeled datasets—a persistent limitation in medical imaging—by producing realistic and diverse lesion-focused samples. This study introduces a generative AI-driven framework for automated brain lesion recognition and classification, comprising standardized preprocessing, targeted data augmentation, and a hybrid discriminative–generative modeling pipeline. The system emphasizes robustness and clinical transparency through explainable inference modules and integrated uncertainty quantification. Experimental results demonstrate that generative learning markedly improves segmentation and classification accuracy, particularly in cases involving rare or morphologically ambiguous lesions. These findings support the integration of generative AI as a cornerstone technology for next-generation precise, scalable, and non-invasive neurodiagnostic workflows
- 1 October 2023
- Published by Edra SpA in Dental Cadmos
- Vol. 91 (08) , 639-726
Tumefazione del mascellare superiore in paziente con iperparatiroidismo
Un paziente di 55 anni viene riferito, dal proprio odontoiatra curante, alla S.C. Odontostomatologia II dell’ASST Santi Paolo e Carlo di Milano per una lesione ulcerativa sospetta in corrispondenza del IV quadrante. All’anamnesi patologica non si evidenziano patologie attuali/pregresse; nessuna terapia farmacologica in atto. Il paziente riferisce di avere eseguito, 7 mesi prima, l’estrazione degli … Continued
- 1 October 2023
- Published by Edra SpA in Dental Cadmos
- Vol. 91 (08) , 639-726
Tumefazione del mascellare superiore in paziente con iperparatiroidismo
Un paziente di 55 anni viene riferito, dal proprio odontoiatra curante, alla S.C. Odontostomatologia II dell’ASST Santi Paolo e Carlo di Milano per una lesione ulcerativa sospetta in corrispondenza del IV quadrante. All’anamnesi patologica non si evidenziano patologie attuali/pregresse; nessuna terapia farmacologica in atto. Il paziente riferisce di avere eseguito, 7 mesi prima, l’estrazione degli … Continued
- Open Access
- 1 January 2016
- Published by Institution of Engineering and Technology (IET) in IET Science, Measurement & Technology
- Vol. 10 (1) , 18-27
Validation of an automatic segmentation method to detect vertebral interfaces in ultrasound images
Aim of this study was to perform a detailed clinical validation of a new fully automatic algorithm for vertebral interface segmentation in echographic images. Abdominal echographic scans of lumbar vertebrae L1–L4 were carried out on 150 female subjects with variable age and body mass index (BMI). Acquired datasets were automatically processed by the algorithm and the accuracy of the obtained segmentations was then evaluated by three independent experienced operators. Obtained results showed a very good specificity in vertebra detection (93.3%), coupled with a reasonable sensitivity (68.1%), representing a suitable compromise between the detection of a sufficient number of vertebrae for reliable diagnoses and the limitation of the corresponding computation time. Importantly, there was only a minimum presence of ‘false vertebrae’ detected (2.8%), resulting in a very low influence on subsequent diagnostic analyses. Furthermore, the algorithm was specifically tuned to provide an improved sensitivity (up to 73.1%) with increasing patient BMI, to keep a suitable number of correctly detected vertebrae even when the acquisition was intrinsically more difficult because of the augmented thickness of abdominal soft tissues. The proposed algorithm will represent an essential added value for developing echographic methods for the diagnosis of osteoporosis on lumbar vertebrae.
- 1 January 2016
- Published by Institution of Engineering and Technology (IET) in IET Science, Measurement & Technology
- Vol. 10 (1) , 18-27
Open AccessValidation of an automatic segmentation method to detect vertebral interfaces in ultrasound images
Aim of this study was to perform a detailed clinical validation of a new fully automatic algorithm for vertebral interface segmentation in echographic images. Abdominal echographic scans of lumbar vertebrae L1–L4 were carried out on 150 female subjects with variable age and body mass index (BMI). Acquired datasets were automatically processed by the algorithm and the accuracy of the obtained segmentations was then evaluated by three independent experienced operators. Obtained results showed a very good specificity in vertebra detection (93.3%), coupled with a reasonable sensitivity (68.1%), representing a suitable compromise between the detection of a sufficient number of vertebrae for reliable diagnoses and the limitation of the corresponding computation time. Importantly, there was only a minimum presence of ‘false vertebrae’ detected (2.8%), resulting in a very low influence on subsequent diagnostic analyses. Furthermore, the algorithm was specifically tuned to provide an improved sensitivity (up to 73.1%) with increasing patient BMI, to keep a suitable number of correctly detected vertebrae even when the acquisition was intrinsically more difficult because of the augmented thickness of abdominal soft tissues. The proposed algorithm will represent an essential added value for developing echographic methods for the diagnosis of osteoporosis on lumbar vertebrae.
- 1 January 2015
- Published by Elsevier in Ultrasound in Medicine & Biology
- Vol. 41 (1) , 281-300
A Novel Ultrasound Methodology for Estimating Spine Mineral Density
- 1 January 2015
- Published by Elsevier in Ultrasound in Medicine & Biology
- Vol. 41 (1) , 281-300
A Novel Ultrasound Methodology for Estimating Spine Mineral Density
- 1 June 2014
- Published by Elsevier in Annals of the Rheumatic Diseases
- Vol. 73 (Suppl 2) , 770-771
SAT0491 Accuracy of A New Ultrasonic Method for Osteoporosis Diagnosis on Lumbar Spine
Background Osteoporosis is a systemic skeletal disease characterized by low bone mass and microarchitectural deterioration of bone tissue with a consequent increase in bone fragility and susceptibility to fracture. Fractures associated with osteoporosis represent a major cause of public health burden in terms of mortality, disability and health care costs emphasizing the need of more effective diagnostic methods. Currently, osteoporosis is mainly diagnosed through dual-energy X-ray absorptiometry (DXA) which provides quantitative measurements of bone mineral density (BMD). Spinal DXA, in particular, is the preferred technique for temporal monitoring of BMD variations, since it has been demonstrated that BMD measurement on this site is the most reliable to predict the global fracture risk. However, DXA cannot be used for early diagnoses through population mass screenings because of issues related to ionizing radiation employment. Objectives Aim of this work is to test the diagnostic accuracy of a novel ultrasound (US)-based method to perform spinal densitometry without employing X-rays. Methods A cohort of 358 female patients was recruited according to the following criteria: 45-65 years, body mass index (BMI) 2, absence of significant severe deambulation impairments, medical prescription for a spinal DXA, signed informed consent. All the enrolled patients underwent two examinations: a conventional spinal DXA (Hologic Discovery) and an US scan of lumbar spine. US data were analyzed by a novel algorithm that processed both echographic images and “raw” radiofrequency signals and, performing a series of spectral and statistical analyses, calculated the same diagnostic parameters provided by DXA (bone mineral density (BMD), T-score, Z-score). Diagnostic accuracy of obtained results was evaluated through a direct comparison with DXA output as a function of patient age. Results For 83.0% of the patients US diagnosis (osteoporotic, osteopenic, healthy) was the same of the corresponding DXA one. In particular, diagnostic accuracy showed the following trend as a function of patient age range: accuracy was 86.4% in 45-50 y, 85.6% in 50-55 y, 87.1% in 55-60 y and 72.4% in 60-65 y. For patients in the same age range, Pearson correlation coefficient (r) between DXA and US measurements was also evaluated for each diagnostic parameter (BMD, T-score, Z-score): all the obtained values of r were within the interval 0.52-0.78 (pConclusions The proposed method for US evaluation of BMD directly on the spine showed a very good and significant agreement with DXA diagnoses. This innovative non-ionizing approach has the potential to anticipate osteoporosis diagnosis by several years through extended screenings in younger populations and to become a useful tool for therapeutic outcome monitoring. Acknowledgements This work was partially funded by FESR P.O. Apulia Region 2007-2013 – Action 1.2.4, grant n. 3Q5AX31. Disclosure of Interest None declared DOI 10.1136/annrheumdis-2014-eular.5446
- 1 June 2014
- Published by Elsevier in Annals of the Rheumatic Diseases
- Vol. 73 (Suppl 2) , 770-771
SAT0491 Accuracy of A New Ultrasonic Method for Osteoporosis Diagnosis on Lumbar Spine
Background Osteoporosis is a systemic skeletal disease characterized by low bone mass and microarchitectural deterioration of bone tissue with a consequent increase in bone fragility and susceptibility to fracture. Fractures associated with osteoporosis represent a major cause of public health burden in terms of mortality, disability and health care costs emphasizing the need of more effective diagnostic methods. Currently, osteoporosis is mainly diagnosed through dual-energy X-ray absorptiometry (DXA) which provides quantitative measurements of bone mineral density (BMD). Spinal DXA, in particular, is the preferred technique for temporal monitoring of BMD variations, since it has been demonstrated that BMD measurement on this site is the most reliable to predict the global fracture risk. However, DXA cannot be used for early diagnoses through population mass screenings because of issues related to ionizing radiation employment. Objectives Aim of this work is to test the diagnostic accuracy of a novel ultrasound (US)-based method to perform spinal densitometry without employing X-rays. Methods A cohort of 358 female patients was recruited according to the following criteria: 45-65 years, body mass index (BMI) 2, absence of significant severe deambulation impairments, medical prescription for a spinal DXA, signed informed consent. All the enrolled patients underwent two examinations: a conventional spinal DXA (Hologic Discovery) and an US scan of lumbar spine. US data were analyzed by a novel algorithm that processed both echographic images and “raw” radiofrequency signals and, performing a series of spectral and statistical analyses, calculated the same diagnostic parameters provided by DXA (bone mineral density (BMD), T-score, Z-score). Diagnostic accuracy of obtained results was evaluated through a direct comparison with DXA output as a function of patient age. Results For 83.0% of the patients US diagnosis (osteoporotic, osteopenic, healthy) was the same of the corresponding DXA one. In particular, diagnostic accuracy showed the following trend as a function of patient age range: accuracy was 86.4% in 45-50 y, 85.6% in 50-55 y, 87.1% in 55-60 y and 72.4% in 60-65 y. For patients in the same age range, Pearson correlation coefficient (r) between DXA and US measurements was also evaluated for each diagnostic parameter (BMD, T-score, Z-score): all the obtained values of r were within the interval 0.52-0.78 (pConclusions The proposed method for US evaluation of BMD directly on the spine showed a very good and significant agreement with DXA diagnoses. This innovative non-ionizing approach has the potential to anticipate osteoporosis diagnosis by several years through extended screenings in younger populations and to become a useful tool for therapeutic outcome monitoring. Acknowledgements This work was partially funded by FESR P.O. Apulia Region 2007-2013 – Action 1.2.4, grant n. 3Q5AX31. Disclosure of Interest None declared DOI 10.1136/annrheumdis-2014-eular.5446
- 1 June 2014
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-5
Fully automatic 3D segmentation measurements of human liver vessels from contrast-enhanced CT
Aim of the present work was to evaluate the performance of a novel fully automatic algorithm for 3D segmentation and volumetric reconstruction of liver vessel network from contrast-enhanced computed tomography (CECT) datasets acquired during routine clinical activity. Three anonymized CECT datasets were randomly collected and were automatically analyzed by the new vessel segmentation algorithm, whose parameter configuration had been previously optimized on a phantom model. The same datasets were also manually segmented by an experienced operator that was blind with respect to algorithm outcome. Automatic segmentation accuracy was quantitatively assessed for both single 2D slices and 3D reconstruction of the vessel network, accounting manual segmentation results as the reference “ground truth”. Adopted evaluation framework included the following two groups of calculations: 1) for 3D vessel network, sensitivity in vessel detection was quantified as a function of both vessel diameter and vessel order; 2) for vessel images on 2D slices, dice similarity coefficient (DSC), false positive ratio (FPR), false negative ratio (FNR), Bland-Altman plots and Pearson correlation coefficients were used to judge the correctness of single pixel classifications. Automatic segmentation resulted in a 3D vessel detection sensitivity of 100% for vessels larger than 1 mm in diameter, 64.6% for vessels in the range 0.5-1.0 mm and 27.8% for smaller vessels. An average area overlap of 99.1% was obtained between automatically and manually segmented vessel sections, with an average difference of 0.53 mm 2 . The corresponding average values of FPR and FNR were 1.8% and 1.6%, respectively. Therefore, the tested method showed significant robustness and accuracy in automatic extraction of the liver vessel tree from CECT datasets. Although further verification studies on larger patient populations are required, the described algorithm has an exciting potential for supporting liver surgery planning and intraoperative resection guidance.
- 1 June 2014
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-5
Fully automatic 3D segmentation measurements of human liver vessels from contrast-enhanced CT
Aim of the present work was to evaluate the performance of a novel fully automatic algorithm for 3D segmentation and volumetric reconstruction of liver vessel network from contrast-enhanced computed tomography (CECT) datasets acquired during routine clinical activity. Three anonymized CECT datasets were randomly collected and were automatically analyzed by the new vessel segmentation algorithm, whose parameter configuration had been previously optimized on a phantom model. The same datasets were also manually segmented by an experienced operator that was blind with respect to algorithm outcome. Automatic segmentation accuracy was quantitatively assessed for both single 2D slices and 3D reconstruction of the vessel network, accounting manual segmentation results as the reference “ground truth”. Adopted evaluation framework included the following two groups of calculations: 1) for 3D vessel network, sensitivity in vessel detection was quantified as a function of both vessel diameter and vessel order; 2) for vessel images on 2D slices, dice similarity coefficient (DSC), false positive ratio (FPR), false negative ratio (FNR), Bland-Altman plots and Pearson correlation coefficients were used to judge the correctness of single pixel classifications. Automatic segmentation resulted in a 3D vessel detection sensitivity of 100% for vessels larger than 1 mm in diameter, 64.6% for vessels in the range 0.5-1.0 mm and 27.8% for smaller vessels. An average area overlap of 99.1% was obtained between automatically and manually segmented vessel sections, with an average difference of 0.53 mm 2 . The corresponding average values of FPR and FNR were 1.8% and 1.6%, respectively. Therefore, the tested method showed significant robustness and accuracy in automatic extraction of the liver vessel tree from CECT datasets. Although further verification studies on larger patient populations are required, the described algorithm has an exciting potential for supporting liver surgery planning and intraoperative resection guidance.
- 1 July 2013
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1809-1812
A new ultrasonic method for lumbar spine densitometry
Aim of this work was to carry out a first clinical validation of a new ultrasound (US)-based approach to bone densitometry of lumbar spine. A total of 290 female patients were enrolled for this study (45-75 years of age, body mass index (BMI)<;40 kg/m 2 ) and all of them underwent two different diagnostic investigations: a lumbar DXA (dual-energy X-ray absorptiometry) and an US scan of the same vertebras, performed with an echographic device configured for the acquisition of both echographic images and unfiltered radiofrequency signals. US data analysis was carried out through an innovative algorithm, whose main features include: a) measurements are always performed on a specific region of interest of the vertebra, identified on the basis of both morphologic and spectral characteristics; b) analysis takes into account patient BMI; c) the algorithm is integrated with a reference database containing model acquisitions for different combinations of patient age, sex and BMI. Accuracy of final algorithm output, represented by the same diagnostic parameters of a DXA investigation, was evaluated through a direct comparison with DXA results. For 84.5% of the patients US diagnosis (osteoporotic, osteopenic, healthy) coincided with the corresponding DXA one and this accuracy level was not appreciably influenced by patient age nor by BMI. The proposed approach represents the first US method for osteoporosis diagnosis which is directly applicable on spine and has the potential to be effectively used for population mass screenings.
- 1 July 2013
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1809-1812
A new ultrasonic method for lumbar spine densitometry
Aim of this work was to carry out a first clinical validation of a new ultrasound (US)-based approach to bone densitometry of lumbar spine. A total of 290 female patients were enrolled for this study (45-75 years of age, body mass index (BMI)<;40 kg/m 2 ) and all of them underwent two different diagnostic investigations: a lumbar DXA (dual-energy X-ray absorptiometry) and an US scan of the same vertebras, performed with an echographic device configured for the acquisition of both echographic images and unfiltered radiofrequency signals. US data analysis was carried out through an innovative algorithm, whose main features include: a) measurements are always performed on a specific region of interest of the vertebra, identified on the basis of both morphologic and spectral characteristics; b) analysis takes into account patient BMI; c) the algorithm is integrated with a reference database containing model acquisitions for different combinations of patient age, sex and BMI. Accuracy of final algorithm output, represented by the same diagnostic parameters of a DXA investigation, was evaluated through a direct comparison with DXA results. For 84.5% of the patients US diagnosis (osteoporotic, osteopenic, healthy) coincided with the corresponding DXA one and this accuracy level was not appreciably influenced by patient age nor by BMI. The proposed approach represents the first US method for osteoporosis diagnosis which is directly applicable on spine and has the potential to be effectively used for population mass screenings.
- 1 June 2013
- Published by Elsevier in Annals of the Rheumatic Diseases
- Vol. 72 (Suppl 3) , A976-A977
AB0612 Evaluation of bone mineral density on femoral neck: preliminary clinical validation of a new ultrasonic method
Background Hip fractures are a major cause of public health burden associated with osteoporosis in terms of mortality, disability and health care costs. The only possible way to reduce the occurrence of hip fractures is represented by the adoption of more effective strategies for osteoporosis diagnosis and hip fracture prevention through population mass screenings. “Gold standard” method for osteoporosis diagnosis is currently represented by measurement of bone mineral density (BMD) obtained through dual-energy X-ray absorptiometry (DXA) and it has been demostrated that BMD measurements on proximal femur provide the most reliable predictions of hip fracture risk. However, because of high costs and issues related to X-ray employment, DXA cannot be used for screening purposes. Ultrasound (US) systems for bone status assessments have been also developed, but all the commercially-available devices are dedicated to peripheral bone districts and their results present poor correlations with femoral neck BMD. Objectives Aim of this work is to carry out a preliminary clinical validation of a novel ultrasonic methodology for bone densitometry directly applicable on femoral neck and capable of providing significant agreement with DXA results. Methods A cohort of 100 female patients was recruited according to the following enrollment criteria: 60-80 years of age, body mass index (BMI) ≤ 40 kg/m2, absence of important deambulation impairments, medical prescription for a femoral DXA, signed informed consent. All the enrolled patients underwent two different diagnostic investigations: a conventional DXA densitometry of the proximal femur (Hologic Discovery) and an US scan of the same bone district, acquiring both echographic images and unfiltered “raw” signals. US data were analyzed by a new algorithm that, through a series of spectral and statistical analyses on both the echographic images and the corresponding unfiltered “raw” signals, and taking into account for each patient the specific combination of age and BMI, calculated the same diagnostic parameters obtained from DXA examination (BMD, T-score, Z-score). Accuracy of each parameter calculated by this algorithm was then evaluated through a direct comparison with DXA results as a function of both patient age and BMI. Results For 81.0% of the studied patients US diagnosis (healthy, osteopenic or osteoporotic patient) was the same of the corresponding DXA one. Pearson correlation coefficient (r) between DXA and US measurements was evaluated for each diagnostic parameter: r=0.68 for BMD, r=0.68 for T-score and r=0.70 for Z-score. Analyzing the correlation values for each group of patients having both BMI and age in the same ranges, we found maximum correlation (r=0.80) for T-score and Z-score values measured on patients with both age in the range 71-75 yr and BMI in 30-40 kg/m2, while minimum correlation (r=0.64) was found for BMD and T-score values measured on patients having both age in the range 61-65 yr and BMI in 25-30 kg/m2. Conclusions The adopted US method for femoral bone densitometry showed an accuracy level comparable with DXA one, in terms of both diagnosis agreement and correlation of single parameter values, therefore showing a great potential for early osteoporosis diagnosis and hip fracture prevention through population mass screenings. Acknowledgements This work was partially funded by FESR P.O. Apulia Region 2007-2013 – Action 1.2.4, grant n. 3Q5AX31. Disclosure of Interest F. Conversano Shareholder of: Echolight srl, a National Research Council spin-off that may or may not benefit from results of this study, E. Casciaro Shareholder of: Echolight srl, a National Research Council spin-off that may or may not benefit from results of this study, A. Greco: None Declared, P. Pisani: None Declared, R. Franchini: None Declared, E. Quarta: None Declared, M. Muratore: None Declared, S. Casciaro Shareholder of: Echolight srl, a National Research Council spin-off that may or may not benefit from results of this study
- 1 June 2013
- Published by Elsevier in Annals of the Rheumatic Diseases
- Vol. 72 (Suppl 3) , A976-A977
AB0612 Evaluation of bone mineral density on femoral neck: preliminary clinical validation of a new ultrasonic method
Background Hip fractures are a major cause of public health burden associated with osteoporosis in terms of mortality, disability and health care costs. The only possible way to reduce the occurrence of hip fractures is represented by the adoption of more effective strategies for osteoporosis diagnosis and hip fracture prevention through population mass screenings. “Gold standard” method for osteoporosis diagnosis is currently represented by measurement of bone mineral density (BMD) obtained through dual-energy X-ray absorptiometry (DXA) and it has been demostrated that BMD measurements on proximal femur provide the most reliable predictions of hip fracture risk. However, because of high costs and issues related to X-ray employment, DXA cannot be used for screening purposes. Ultrasound (US) systems for bone status assessments have been also developed, but all the commercially-available devices are dedicated to peripheral bone districts and their results present poor correlations with femoral neck BMD. Objectives Aim of this work is to carry out a preliminary clinical validation of a novel ultrasonic methodology for bone densitometry directly applicable on femoral neck and capable of providing significant agreement with DXA results. Methods A cohort of 100 female patients was recruited according to the following enrollment criteria: 60-80 years of age, body mass index (BMI) ≤ 40 kg/m2, absence of important deambulation impairments, medical prescription for a femoral DXA, signed informed consent. All the enrolled patients underwent two different diagnostic investigations: a conventional DXA densitometry of the proximal femur (Hologic Discovery) and an US scan of the same bone district, acquiring both echographic images and unfiltered “raw” signals. US data were analyzed by a new algorithm that, through a series of spectral and statistical analyses on both the echographic images and the corresponding unfiltered “raw” signals, and taking into account for each patient the specific combination of age and BMI, calculated the same diagnostic parameters obtained from DXA examination (BMD, T-score, Z-score). Accuracy of each parameter calculated by this algorithm was then evaluated through a direct comparison with DXA results as a function of both patient age and BMI. Results For 81.0% of the studied patients US diagnosis (healthy, osteopenic or osteoporotic patient) was the same of the corresponding DXA one. Pearson correlation coefficient (r) between DXA and US measurements was evaluated for each diagnostic parameter: r=0.68 for BMD, r=0.68 for T-score and r=0.70 for Z-score. Analyzing the correlation values for each group of patients having both BMI and age in the same ranges, we found maximum correlation (r=0.80) for T-score and Z-score values measured on patients with both age in the range 71-75 yr and BMI in 30-40 kg/m2, while minimum correlation (r=0.64) was found for BMD and T-score values measured on patients having both age in the range 61-65 yr and BMI in 25-30 kg/m2. Conclusions The adopted US method for femoral bone densitometry showed an accuracy level comparable with DXA one, in terms of both diagnosis agreement and correlation of single parameter values, therefore showing a great potential for early osteoporosis diagnosis and hip fracture prevention through population mass screenings. Acknowledgements This work was partially funded by FESR P.O. Apulia Region 2007-2013 – Action 1.2.4, grant n. 3Q5AX31. Disclosure of Interest F. Conversano Shareholder of: Echolight srl, a National Research Council spin-off that may or may not benefit from results of this study, E. Casciaro Shareholder of: Echolight srl, a National Research Council spin-off that may or may not benefit from results of this study, A. Greco: None Declared, P. Pisani: None Declared, R. Franchini: None Declared, E. Quarta: None Declared, M. Muratore: None Declared, S. Casciaro Shareholder of: Echolight srl, a National Research Council spin-off that may or may not benefit from results of this study
- 1 June 2013
- Published by Elsevier in Annals of the Rheumatic Diseases
- Vol. 72 (Suppl 3) , A980
AB0624 High correlation between a new ultrasonic method for spinal densitometry and dual x-ray absorptiometry
Background Osteoporosis is a systemic skeletal disease characterized by low bone mass and microarchitectural deterioration of bone tissue leading to increased fragility and augmented fracture risk. Bone densitometry performed through dual X-ray absorptiometry (DXA) provides quantitative measurements of bone mineral density (BMD) and such measurements currently represent the reference indicated by World Health Organization for osteoporosis diagnosis. The most frequently investigated anatomical sites are lumbar spine and proximal femur, since it has been demonstrated that BMD measurements on these sites are the most reliable to predict the global fracture risk. Spinal DXA, in particular, is the preferred technique for temporal monitoring of BMD variations. However, DXA examinations exploit X-rays and for this reason they cannot be used for population mass screenings aimed at providing early osteoporosis diagnosis. Objectives Aim of this work is to perform a preliminary clinical validation of a new technique for spinal bone densitometry, which does not require ionizing radiation but employs only ultrasound (US) signals. Methods A total of 280 female patients were recruited for this study according to the following criteria: 45-75 years of age, body mass index (BMI) ≤ 40 kg/m2, absence of significant deambulation impairments, medical prescription for a spinal DXA, signed informed consent. All the enrolled patients underwent two different diagnostic examinations: a conventional spinal DXA (Hologic Discovery) and an US scan of lumbar spine, acquiring both echographic images and unfiltered “raw” signals. US data were analyzed through a new algorithm that performed a series of spectral and statistical analyses, involving both the echographic images and the corresponding unfiltered “raw” US signals, and provided as final output the same diagnostic parameters obtained from a DXA examination (BMD, T-score, Z-score). Accuracy of each parameter calculated by this algorithm was then evaluated through a direct comparison with DXA results as a function of both patient age and BMI. Results For 86.1% of the studied patients US diagnosis (healthy, osteopenic or osteoporotic patient) was the same of the corresponding DXA one. In particular, diagnostic accuracy was 86.7% for patients with BMI2 (n=158), 89.2% for those with BMI in the range 25-30 kg/m2 (n=83) and 76.9% for those with BMI in the range 30-40 kg/m2 (n=39). Assessing diagnostic accuracy as a function of patient age, we obtained the following results: 84.1% for patients in the age range 45-55 yr (n=69), 87.5% in 56-65 yr (n=160), 84.3% in 66-75 yr (n=51). For each group of patients having both BMI and age in the same ranges, Pearson correlation coefficient (r) between DXA and US results was calculated for each diagnostic parameter (BMD, T-score, Z-score): all the obtained values were in the range 0.67-0.88 (pConclusions The proposed method represents the first US technique for BMD assessment directly on spine. Reported results showed an unprecedented agreement between DXA and US diagnosis and are therefore encouraging in view of future applications of this method for early diagnosis of osteoporosis and therapeutic monitoring. Acknowledgements This work was partially funded by FESR P.O. Apulia Region 2007-2013 – Action 1.2.4, grant n. 3Q5AX31. Disclosure of Interest M. Muratore: None Declared, F. Conversano Shareholder of: Echolight srl, a National Research Council spin-off that may or may not benefit from results of this study, E. Casciaro Shareholder of: Echolight srl, a National Research Council spin-off that may or may not benefit from results of this study, G. Soloperto: None Declared, R. Franchini: None Declared, A. Greco: None Declared, E. Quarta: None Declared, A. Grimaldi: None Declared, S. Casciaro Shareholder of: Echolight srl, a National Research Council spin-off that may or may not benefit from results of this study
- 1 June 2013
- Published by Elsevier in Annals of the Rheumatic Diseases
- Vol. 72 (Suppl 3) , A980
AB0624 High correlation between a new ultrasonic method for spinal densitometry and dual x-ray absorptiometry
Background Osteoporosis is a systemic skeletal disease characterized by low bone mass and microarchitectural deterioration of bone tissue leading to increased fragility and augmented fracture risk. Bone densitometry performed through dual X-ray absorptiometry (DXA) provides quantitative measurements of bone mineral density (BMD) and such measurements currently represent the reference indicated by World Health Organization for osteoporosis diagnosis. The most frequently investigated anatomical sites are lumbar spine and proximal femur, since it has been demonstrated that BMD measurements on these sites are the most reliable to predict the global fracture risk. Spinal DXA, in particular, is the preferred technique for temporal monitoring of BMD variations. However, DXA examinations exploit X-rays and for this reason they cannot be used for population mass screenings aimed at providing early osteoporosis diagnosis. Objectives Aim of this work is to perform a preliminary clinical validation of a new technique for spinal bone densitometry, which does not require ionizing radiation but employs only ultrasound (US) signals. Methods A total of 280 female patients were recruited for this study according to the following criteria: 45-75 years of age, body mass index (BMI) ≤ 40 kg/m2, absence of significant deambulation impairments, medical prescription for a spinal DXA, signed informed consent. All the enrolled patients underwent two different diagnostic examinations: a conventional spinal DXA (Hologic Discovery) and an US scan of lumbar spine, acquiring both echographic images and unfiltered “raw” signals. US data were analyzed through a new algorithm that performed a series of spectral and statistical analyses, involving both the echographic images and the corresponding unfiltered “raw” US signals, and provided as final output the same diagnostic parameters obtained from a DXA examination (BMD, T-score, Z-score). Accuracy of each parameter calculated by this algorithm was then evaluated through a direct comparison with DXA results as a function of both patient age and BMI. Results For 86.1% of the studied patients US diagnosis (healthy, osteopenic or osteoporotic patient) was the same of the corresponding DXA one. In particular, diagnostic accuracy was 86.7% for patients with BMI2 (n=158), 89.2% for those with BMI in the range 25-30 kg/m2 (n=83) and 76.9% for those with BMI in the range 30-40 kg/m2 (n=39). Assessing diagnostic accuracy as a function of patient age, we obtained the following results: 84.1% for patients in the age range 45-55 yr (n=69), 87.5% in 56-65 yr (n=160), 84.3% in 66-75 yr (n=51). For each group of patients having both BMI and age in the same ranges, Pearson correlation coefficient (r) between DXA and US results was calculated for each diagnostic parameter (BMD, T-score, Z-score): all the obtained values were in the range 0.67-0.88 (pConclusions The proposed method represents the first US technique for BMD assessment directly on spine. Reported results showed an unprecedented agreement between DXA and US diagnosis and are therefore encouraging in view of future applications of this method for early diagnosis of osteoporosis and therapeutic monitoring. Acknowledgements This work was partially funded by FESR P.O. Apulia Region 2007-2013 – Action 1.2.4, grant n. 3Q5AX31. Disclosure of Interest M. Muratore: None Declared, F. Conversano Shareholder of: Echolight srl, a National Research Council spin-off that may or may not benefit from results of this study, E. Casciaro Shareholder of: Echolight srl, a National Research Council spin-off that may or may not benefit from results of this study, G. Soloperto: None Declared, R. Franchini: None Declared, A. Greco: None Declared, E. Quarta: None Declared, A. Grimaldi: None Declared, S. Casciaro Shareholder of: Echolight srl, a National Research Council spin-off that may or may not benefit from results of this study
- 1 May 2013
- Published by Bioscientifica in Bone Abstracts
Strong diagnostic agreement between a novel ultrasound-based method for lumbar densitometry and dual-energy X-ray absorptiometry
Introduction: Currently, osteoporosis is mainly diagnosed through dual-enargy X-ray absorptiometry (DXA). However, DXA cannot be used for early diagnoses through population mass screenings because of issues related to ionizing radiation employment. Aim of this study is to perform a preliminary clinical validation of a new ultrasound (US)-based method for vertebral densitometry. Methods: A total of 270 women were included in this study according to the following criteria: 45–80 years of age, BMI ≤40 kg/m2, no deambulation impairments, medical prescription for a vertebral DXA, signed informed consent. All the enrolled patients underwent two examinations: a conventional vertebral DXA (Hologic Discovery) and an US scan of lumbar spine. US data were analyzed by a novel algorithm that processed both echographic images and corresponding unfiltered ‘raw’ signals and calculated the same diagnostic parameters provided by DXA (bone mineral density (BMD), T-score, Z-score). Diagnostic accuracy of obtained results was assessed through a direct comparison with DXA output as a function of patient age and BMI. Results: For 87.0% of the patients US diagnosis (osteoporotic, osteopenic, and healthy) was the same of the corresponding DXA one. Specifically, diagnostic accuracy was 87.7% for patients with BMI in the range 25–40 kg/m2 (n=114) and 86.5% for those with BMI 2 (n=156), with maximum (88.6%) and minimum (78.7%) accuracy in the age range 61–65 and 45–50 years, respectively. All the obtained values of Pearson correlation coefficient (r) between diagnostic parameters provided by DXA and US for patients in the same age and BMI ranges were within the interval 0.72–0.91 (P<0.001). Conclusions: We proposed an innovative method for US evaluation of BMD directly on the spine which showed a strong and significant agreement with DXA diagnoses. This technique has the potential to revolutionize the approach to osteoporosis diagnosis.
- 1 May 2013
- Published by Bioscientifica in Bone Abstracts
Strong diagnostic agreement between a novel ultrasound-based method for lumbar densitometry and dual-energy X-ray absorptiometry
Introduction: Currently, osteoporosis is mainly diagnosed through dual-enargy X-ray absorptiometry (DXA). However, DXA cannot be used for early diagnoses through population mass screenings because of issues related to ionizing radiation employment. Aim of this study is to perform a preliminary clinical validation of a new ultrasound (US)-based method for vertebral densitometry. Methods: A total of 270 women were included in this study according to the following criteria: 45–80 years of age, BMI ≤40 kg/m2, no deambulation impairments, medical prescription for a vertebral DXA, signed informed consent. All the enrolled patients underwent two examinations: a conventional vertebral DXA (Hologic Discovery) and an US scan of lumbar spine. US data were analyzed by a novel algorithm that processed both echographic images and corresponding unfiltered ‘raw’ signals and calculated the same diagnostic parameters provided by DXA (bone mineral density (BMD), T-score, Z-score). Diagnostic accuracy of obtained results was assessed through a direct comparison with DXA output as a function of patient age and BMI. Results: For 87.0% of the patients US diagnosis (osteoporotic, osteopenic, and healthy) was the same of the corresponding DXA one. Specifically, diagnostic accuracy was 87.7% for patients with BMI in the range 25–40 kg/m2 (n=114) and 86.5% for those with BMI 2 (n=156), with maximum (88.6%) and minimum (78.7%) accuracy in the age range 61–65 and 45–50 years, respectively. All the obtained values of Pearson correlation coefficient (r) between diagnostic parameters provided by DXA and US for patients in the same age and BMI ranges were within the interval 0.72–0.91 (P<0.001). Conclusions: We proposed an innovative method for US evaluation of BMD directly on the spine which showed a strong and significant agreement with DXA diagnoses. This technique has the potential to revolutionize the approach to osteoporosis diagnosis.