Advanced Techniques for Face Recognition under Challenging Environments

Automatically recognizing faces captured under uncontrolled environments has always been a challenging topic in the past decades. In this work, we investigate cohort score normalization that has been widely used in biometric verification as means to improve the robustness of face recognition under challenging environments. In particular, we introduce cohort score normalization into undersampled face recognition problem. Further, we develop an effective cohort normalization method specifically for the unconstrained face pair matching problem. Extensive experiments conducted on several well known face databases demonstrate the effectiveness of cohort normalization on these challenging scenarios. In addition, to give a proper understanding of cohort behavior, we study the impact of the number and quality of cohort samples on the normalization performance. The experimental results show that bigger cohort set size gives more stable and often better results to a point before the performance saturates. And cohort samples with different quality indeed produce different cohort normalization performance. Recognizing faces gone after alterations is another challenging problem for current face recognition algorithms. Face image alterations can be roughly classified into two categories: unintentional (e.g., geometrics transformations introduced by the acquisition devide) and intentional alterations (e.g., plastic surgery). We study the impact of these alterations on face recognition accuracy. Our results show that state-of-the-art algorithms are able to overcome limited digital alterations but are sensitive to more relevant modifications. Further, we develop two useful descriptors for detecting those alterations which can significantly affect the recognition performance. In the end, we propose to use the Structural Similarity (SSIM) quality map to detect and model variations due to plastic surgeries. Extensive experiments conducted on a plastic surgery face database demonstrate the potential of SSIM map for matching face images after surgeries.

Anatomic study of the feline bronchial and vascular structures with a 64 detector-row computed tomography

Multidetector row computed tomography over the last decade is commonly used in veterinary medicine. This new technology has an increased spatial and temporal resolution, could evaluate wider scanning range in shorter scanning time, providing an advanced imaging modality. Computed tomography angiographic studies are commonly used in veterinary medicine in order to evaluate vascular structures of the abdomen and the thorax. Pulmonary pathology in feline patients is a very common condition and usually is further evaluating with computed tomography. Up to date few references of the normal computed tomographic aspects of the feline thorax are reported. In this study a computed tomographic pulmonary angiography (CTPA) protocol is reported in normal cats and is compared with the up to date anatomical references. A CTPA protocol using a 64 MDCT in our study achieved high resolution images of the pulmonary arteries, pulmonary veins and bronchial lumen till the level of minor segmental branches. Feline pulmonary bronchial parenchyma demonstrates an architecture of mixed type with a monopedial model observed in the most anatomical parts and the dichotomic aspect is seen at the accessory lobe. The arterial and venous architecture is similar to the bronchial. Statistical analysis demonstrates the linear correlation of tracheal diameter to the felines weight. Vascular variations were noticed. The pulmonary venous system enters into the left atrium through three ostia (left cranial ostia: consisted of the anastomosis of the cranial and caudal portion of the left cranial pulmonary vein; right ostia: consisted of the anastomosis of the right cranial and middle pulmonary vein; and the caudal ostia: consisted of the anastomosis of the right and left caudal pulmonary vein). In conclusion CTPA is applicable in feline patients and provides an excellent imaging of the pulmonary arterial, venous and bronchial system till the level of minor segmental branches.

Indicazione e risultati dei trattamenti locoregionali nell’epatocarcinoma

La chemioembolizzazione (TACE) è uno dei trattamenti locoregionali più largamente utilizzati nel trattamento dell’epatocarcinoma (HCC). A tutt’oggi però rimangono irrisolte alcune importanti controversie sul suo impiego. Nella presente tesi sono stati analizzati alcuni dei principali oggetti di dibattito quali (1) indicazione al trattamento, (2) trattamenti multipli e schema di ritrattamento e (3) trattamento dei pazienti candidabili a trapianto di fegato. A tal fine sono stati riportati tre studi che hanno analizzato gli argomenti sopradescritti. La TACE viene comunemente eseguita nei pazienti al di fuori delle raccomandazioni delle linee guida tra cui i pazienti con nodulo singolo, i pazienti con trombosi portale e con performance status (PS) compromesso. Dallo studio 1 è emerso che la TACE può essere considerata una valida opzione terapeutica nei pazienti con HCC singolo non candidabili a trattamenti curativi, che la trombosi portale non neoplastica ed una lieve compromissione del performance status (PS-1) verosimilmente legata alla cirrosi non hanno impatto sulla sopravvivenza post-trattamento. Multipli trattamenti di chemioembolizzazione vengono frequentemente eseguiti ma non esiste a tutt’oggi un numero ottimale di ritrattamenti TACE. Dallo studio 2 è emerso che il trattamento TACE eseguito “on demand” può essere efficacemente ripetuto nei pazienti che non abbiano scompenso funzionale e non siano candidabili a trattamenti curativi anche se solo una piccola percentuale di pazienti selezionati può essere sottoposto a più cicli di trattamento. La TACE è frequentemente impiegata nei pazienti in lista per trapianto di fegato ma non c’è evidenza dell’efficacia di trattamenti ripetuti in questi pazienti. Dallo studio 3 è emerso che il numero di TACE non è significativamente associato né alla necrosi tumorale, né alla recidiva né alla sopravvivenza post-trapianto. Un tempo d’attesa prima del trapianto ≤6 mesi è invece risultato essere fattore predittivo indipendente di recidiva riflettendo la possibile maggiore aggressività tumorale in questa classe di pazienti.

Study on NGF and VEGF during the equine perinatal period

Part 1 of the study aims to: evaluate NGF and VEGF levels obtained at parturition from mare, foal and umbilical cord vein plasma, as well as in amniotic fluid; evaluate NGF and VEGF content in plasma of healthy foals during the first 72 h of life; evaluate NGF and VEGF levels at parturition in relation to selected mares’ and foals’ clinical parameters; evaluate the relationship between the two trophic factors and thyroid hormone levels in the first 72 h of life; assess mRNA expression of NGF, VEGF and BDNF and their cell surface receptors in the placenta. Part 2 aims to clinically characterize a population of foals spontaneously affected by Neonatal Encephalopathy (NE), and then to: evaluate NGF and VEGF levels in plasma samples obtained in the affected population at parturition from mare’s jugular vein, umbilical cord vein and foal’s jugular vein, as well as in amniotic fluid; evaluate NGF and VEGF content in plasma of foals affected by NE during the first 72 h of life/hospitalization; evaluate NGF and VEGF levels at birth/admission in relation to selected mares’ and foals’ clinical parameters; evaluate the relationship between the two trophic factors and thyroid hormone levels in the first 72 h of life/hospitalization; assess the mRNA expression of NGF, VEGF and BDNF, and their cell surface receptors, in the placenta of mares that delivered affected foals. The close relationship between the two trophic factors in foal plasma over time and their fine expression in placental tissues under physiological conditions appear to be key regulators of fetal development and adaptation. Their less pronounced decrease in compromised foals compared to healthy ones, their relationship with thyroid hormones over time, and the reduced expression of NGF and BDNF in placental tissues, could be key regulators in the mechanisms of NE.

Valutazione non invasiva dell'epatopatia Fontan indotta

Introduzione: L’intervento di Fontan comporta un aumento istantaneo della pressione venosa centrale che causa, nel medio-lungo termine, una forma di epatopatia specifica detta FALD. Il monitoraggio della FALD è complesso ma potrebbe consentire di bloccarne o rallentarne l’insorgenza. Lo studio ha valutato l’efficacia delle modalità di monitoraggio non invasivo. Materiale e metodi: Sei pazienti (età media 24 anni) operati presso l’IRCCS Azienda Ospedaliero Universitaria di Bologna sono stati sottoposti a RMN 4D-Flow e ad Ecodoppler epatico. Sono stati raccolti i dati anagrafici, morfologici, anamnestici e i markers sierologici per il calcolo degli scores MELD-XI, APRI, FIB4, i valori di Shear Stress assiale e circonferenziale e gli indici di pulsatilità e resistenza delle arterie epatica e renale. Risultati: Il tempo trascorso tra la Fontan e lo studio è stato di 17,8 anni. Età media alla Fontan 6,8 anni. Tutti i pazienti avevano un quadro compatibile con epatopatia. I markers sierologici e gli scores MELD-XI,APRI e FIB4 si sono dimostrati di scarsa utilità. All’ecografia tutti i pazienti avevano ecostruttura irregolare, splenomegalia e valori elevati di pulsatilità e resistenza dell’arteria epatica e splenica. La rigidità epatica media è stata di 12,4 Kpa. Alla RMN 4DF lo Shear stress assiale è stato massimo a livello del condotto (0,16 Pa) e minimo a livello delle vene sovra epatiche (0,05 Pa). Lo Shear Stress si è mostrato massimo nei pazienti con emodinamica sfavorevole e peggior quadro ecografico addominale, evidenziando aree di inefficienza energetica. Conclusioni: La combinazione delle diagnostiche di imaging non invasive potrebbe rivelarsi adeguata per il monitoraggio della FALD. In particolare, la RMN 4D Flow potrebbe rivelare aree di inefficienza energetica predisponenti alla FALD. Questo potrebbe indirizzare in modo specifico la terapia dei pazienti operati o addirittura indurre la modifica del disegno della Fontan verso forme più efficienti.

Application of augmented reality in craniofacial surgery: a feasibility study

Augmented Reality (AR) is a novel promising technology, which is gaining success in the medical field. A number of applications in surgery have been described, but few studies have been focusing on pediatric craniofacial surgery. In this research project, the Authors have been implementing a system for intraoperative surgical navigation by means of HoloLens 2 by Microsoft, applied to pediatric craniofacial surgery. The Authors tested the device in a preclinical setting first, and then moved to patients. The Authors assessed the accuracy of the HoloLens 2 by performing 36 procedures in vitro on a printed 3D model of a patient. In clinical setting, 10 patients were prospectively enrolled in the study. The virtual surgical planning was designed for each patient and uploaded onto the software which allows for the AR interface and the standard neurosurgical navigator. For each patient, the surgeon has been drawing osteotomy lines both under the guidance of HoloLens2 and of the neurosurgical navigator. The Author then checked the accuracy with calibrated CAD CAM cutting guides with different grooves, in order to assess the accuracy of the osteotomies performed. We tested levels of accuracy of ±1.5 mm and ±1mm . In the preclinical setting, the HoloLens 2 performed with levels of accuracy of 1.5 mm, whereas in the real setting, surgeons were able to trace the osteotomy lines under the AR guidance for an amount of 45% (0.4 SD) of the entire line, with an accuracy level of ±1.5 mm. This percentage lowers to 34% (0.4 SD) when assessing accuracy level of ±1 mm. The results of the same tasks for the standard navigator are 36% and 16%, for ±1.5 mm and ± 1 mm accuracy level, respectively. The Authors reported encouraging results both in the preclinical and the clinical setting.