1,721,074 research outputs found
SCALING ARTIFICIAL INTELLIGENCE IN ENDOSCOPY: FROM MODEL DEVELOPMENT TO MACHINE LEARNING OPERATIONS FRAMEWORKS
Full text linkQuesta tesi esplora l'integrazione dell'intelligenza artificiale (IA) in Otorinolaringoiatria – Chirurgia di Testa e Collo, concentrandosi sui progressi della computer vision per l’endoscopia e le procedure chirurgiche. La ricerca inizia con una revisione completa dello stato dell’arte dell'IA e della computer vision in questo campo, identificando aree per ulteriori sviluppi. L'obiettivo principale è stato quello di sviluppare un sistema di computer vision per l'analisi di immagini e video endoscopici. La ricerca ha coinvolto la progettazione di strumenti per la rilevazione e segmentazione di neoplasie nelle vie aerodigestive superiori (VADS) e la valutazione della motilità delle corde vocali, cruciale nella stadiazione del carcinoma laringeo. Inoltre, lo studio si è focalizzato sul potenziale dei foundation vision models, vision transformers basati su self-supervised learning, per ridurre la necessità di annotazione da parte di esperti, approccio particolarmente vantaggioso in campi con dati limitati. Inoltre, la ricerca ha incluso lo sviluppo di un'applicazione web per migliorare e velocizzare il processo di annotazione in endoscopia delle VADS, nell’ambito generale delle tecniche di MLOps.
La tesi copre varie fasi della ricerca, a partire dalla definizione del quadro concettuale e della metodologia, denominata "Videomics". Include una revisione della letteratura sull'IA in endoscopia clinica, focalizzata sulla Narrow Band Imaging (NBI) e sulle reti neurali convoluzionali (CNN). Lo studio progredisce attraverso diverse fasi, dalla valutazione della qualità delle immagini endoscopiche alla caratterizzazione approfondita delle lesioni neoplastiche. Si affronta anche la necessità di standard nel reporting degli studi di computer vision in ambito medico e si valuta l'applicazione dell'IA in setting dinamici come la motilità delle corde vocali. Una parte significativa della ricerca indaga l'uso di algoritmi di computer vision generalizzati (“foundation models”) e la “commoditization” degli algoritmi di machine learning, utilizzando polipi nasali e il carcinoma orofaringeo come casi studio. Infine, la tesi discute lo sviluppo di ENDO-CLOUD, un sistema basato su cloud per l’analisi della videolaringoscopia, evidenziando le sfide e le soluzioni nella gestione dei dati e l’utilizzo su larga scala di modelli di IA nell'imaging medico.This thesis explores the integration of artificial intelligence (AI) in Otolaryngology – Head and Neck Surgery, focusing on advancements in computer vision for endoscopy and surgical procedures. It begins with a comprehensive review of AI and computer vision advancements in this field, identifying areas for further exploration. The primary aim was to develop a computer vision system for endoscopy analysis. The research involved designing tools for detecting and segmenting neoplasms in the upper aerodigestive tract (UADT) and assessing vocal fold motility, crucial in laryngeal cancer staging.
Further, the study delves into the potential of vision foundation models, like vision transformers trained via self-supervision, to reduce the need for expert annotations, particularly beneficial in fields with limited cases. Additionally, the research includes the development of a web application for enhancing and speeding up the annotation process in UADT endoscopy, under the umbrella of Machine Learning Operations (MLOps).
The thesis covers various phases of research, starting with defining the conceptual framework and methodology, termed "Videomics". It includes a literature review on AI in clinical endoscopy, focusing on Narrow Band Imaging (NBI) and convolutional neural networks (CNNs). The research progresses through different stages, from quality assessment of endoscopic images to in-depth characterization of neoplastic lesions. It also addresses the need for standards in medical computer vision study reporting and evaluates the application of AI in dynamic vision scenarios like vocal fold motility.
A significant part of the research investigates the use of "general purpose" vision algorithms and the commoditization of machine learning algorithms, using nasal polyps and oropharyngeal cancer as case studies. Finally, the thesis discusses the development of ENDO-CLOUD, a cloud-based system for videolaryngoscopy, highlighting the challenges and solutions in data management and the large-scale deployment of AI models in medical imaging
Cervical exenteration and its variants for locally advanced thyroid cancer: when, why, and how?
No full textPurpose of reviewTo describe the modern surgical approach for management of advanced thyroid cancers infiltrating the cervicovisceral axis with special attention to well differentiated tumors not amenable to organ-sparing techniques. In particular, cervical exenteration, herein defined as the sum of total thyroidectomy, central compartment and lateral neck dissections, variously associated with total laryngectomy and possible partial or total pharyngoesophagectomy, represents an extreme surgical procedure that, in properly selected cases, allows for reasonable palliation of central compartment life-threatening signs/symptoms if not cure for an advanced oncologic condition.Recent findingsCervical exenteration is not contraindicated by the presence of limited distant metastases at presentation. Even though it requires that the patient is in general good health as it can be associated with a number of complications and long in-hospital stay, when appropriately planned and performed according to the most recent reconstructive nuances, it allows good oncologic outcomes that are not inferior to those described for similarly advanced primaries of the upper aerodigestive tract. In addition, quality of life and functional results are not significantly different from those described after total laryngectomy for primary laryngeal squamous cell carcinomas.Cervical exenteration requires a tertiary, expert, multidisciplinary effort in terms of diagnosis, surgical performance, and postoperative care. A patient-centered decision process is strongly warranted taking into consideration alternative therapeutic and symptom-based palliative strategies
Videomics: bringing deep learning to diagnostic endoscopy
Full text linkPURPOSE OF REVIEW: Machine learning (ML) algorithms have augmented human judgment in various fields of clinical medicine. However, little progress has been made in applying these tools to video-endoscopy. We reviewed the field of video-analysis (herein termed 'Videomics' for the first time) as applied to diagnostic endoscopy, assessing its preliminary findings, potential, as well as limitations, and consider future developments.RECENT FINDINGS: ML has been applied to diagnostic endoscopy with different aims: blind-spot detection, automatic quality control, lesion detection, classification, and characterization. The early experience in gastrointestinal endoscopy has recently been expanded to the upper aerodigestive tract, demonstrating promising results in both clinical fields. From top to bottom, multispectral imaging (such as Narrow Band Imaging) appeared to provide significant information drawn from endoscopic images.SUMMARY: Videomics is an emerging discipline that has the potential to significantly improve human detection and characterization of clinically significant lesions during endoscopy across medical and surgical disciplines. Research teams should focus on the standardization of data collection, identification of common targets, and optimal reporting. With such a collaborative stepwise approach, Videomics is likely to soon augment clinical endoscopy, significantly impacting cancer patient outcomes
A roadmap of six different pathways to improve survival in laryngeal cancer patients
No full textLaryngeal cancer continues to require improvement in earlier stage diagnosis and better imaging delineation of disease, and hence 'more evidence-based' selection of treatment, as recent evidence suggests that related mortality, in the last decades, has not significantly decreased worldwide. Even though the reasons are not fully understood, there persists an urgency for a review and development of future strategies to embrace such clinical and diagnostic challenges from a political, societal, as well as scientific and clinical points of view
Functional outcomes and donor site morbidity after palatomaxillary reconstruction using the scapular angle osteomuscular free flap.
No full textConservative surgery for laryngeal chondrosarcoma: a review of the most recently proposed approaches
No full textThe aim of this study was to describe the most recent technical nuances for resection and reconstruction of Grade 1 and 2 laryngeal chondrosarcomas, with a special emphasis on those located at the level of the cricoid plate, which is the site of origin of the vast majority of these rare tumours
End-to-end versus end-to-side venous microanastomoses in head and neck reconstruction.
No full textVenous thrombosis (VT) is the primary reason for microvascular free flaps (MFFs) failure. Different series have addressed the influence of venous anastomosis, end-to-end (ETE) vs. end-to-side (ETS), on this issue in head and neck (HN) microsurgery, but a consensus about the optimal technique to be adopted is still lacking. The aim of this study is to prospectively compare the venous complication rates of ETE and ETS techniques in 422 homogeneously treated patients who underwent MFF for HN oncologic defects between 2000 and 2012 at our Institution. Patients were divided into two groups: Group A (n = 269 patients) receiving an ETE and Group B (n = 153) an ETS venous anastomosis. The choice between the type of venous anastomosis was based on the several variables: availability of adequate caliber recipient veins in the neck, length and caliber of the donor vein, geometry and orientation of the vascular pedicle, and possibility to create a tensionless anastomosis. An ETE anastomosis was always preferred when feasible, while an ETS (performed on the internal jugular vein) was reserved to cases in which the abovementioned considerations contraindicated an ETE. Overall, the MFF failure rate was 3 %. Among the 13 failures, five had VT (1.1 %): three had received an ETE, and two an ETS. Venous anastomosis re-exploration and failure rates of the two groups were compared by the Chi-squared test showing no statistically significant differences. In conclusion, our data show how ETS venous anastomosis is a safe alternative to ETE whenever the latter cannot be properly accomplished for the previously mentioned contraindications
Microdebrider cavitation and transcervical removal of parapharyngeal schwannomas approaching the skull base.
No full textRemoval of parapharyngeal space (PPS) schwannomas approaching the skull base through a purely transcervical approach requires adequate visualization of the surgical field to obtain complete resection with minimal sequelae. This is a retrospective series of four patients undergoing transcervical removal of sympathetic chain PPS schwannomas abutting the skull base by an intracapsular microdebrider tumor cavitation. Radiologic data, complications, functional outcomes, and follow-up status were considered. MRI was suggestive of PPS schwannoma in all cases, and correctly predicted the nerve of origin in three out of four cases. All patients developed postoperative Claude Bernard-Horner and first-bite syndromes. One patient also presented temporary neuropraxia of the IX cranial nerve, and another of the IX and X cranial nerves. Microdebrider cavitation of sympathetic chain PPS schwannoma abutting the skull base proved to be a reliable technique allowing good visualization of adjacent neural and vascular structures through a purely transcervical approach, while maintaining a low complication rate
- …
