Performance and clinical implications of machine learning models for detecting cervical ossification of the posterior longitudinal ligament: a systematic review

Asian Spine Journal, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 20, 2025

Ossification of the posterior longitudinal ligament (OPLL) is a significant spinal condition that can lead to severe neurological deficits.Recent advancements in machine learning (ML) and deep (DL) have led development promising tools for early detection diagnosis OPLL.This systematic review evaluated diagnostic performance ML DL models clinical implications OPLL detection.A was conducted following Preferred Reporting Items Systematic Reviews Meta-Analyses guidelines.PubMed/Medline Scopus databases were searched studies published between January 2000 September 2024.Eligible included those utilizing or using imaging data.All assessed risk bias appropriate tools.The key metrics, including accuracy, sensitivity, specificity, area under curve (AUC), analyzed.Eleven studies, comprising total 6,031 patients, included.The demonstrated high performance, with accuracy rates ranging from 69.6% 98.9% AUC values up 0.99.Convolutional neural networks random forest most used approaches.The overall moderate, concerns primarily related participant selection missing data.In conclusion, show great potential accurate OPLL, particularly when integrated techniques.However, ensure applicability, further research warranted validate these findings more extensive diverse populations.

Language: Английский

---

Clinical Outcomes and Patient Perspectives in Full Endoscopic Cervical Surgery: A Systematic Review

Neurospine, Journal Year: 2025, Volume and Issue: 22(1), P. 81 - 104

Published: March 31, 2025

Objective: Full endoscopic cervical surgery (FECS) is an evolving minimally invasive approach for treating spine disorders. This systematic review synthesizes current evidence on the clinical outcomes and patient perspectives associated with FECS, specifically evaluating its safety, efficacy, overall satisfaction.Methods: A search of PubMed/MEDLINE, Cochrane Library, Embase, Web Science databases was conducted following PRISMA (Preferred Reporting Items Systematic Reviews Meta-Analyses) guidelines. Studies published between January 2000 September 2024 that reported or related to FECS were included. Risk bias assessed using ROBINS-I (Risk Of Bias In Non-randomized - Interventions) tool tool. Inclusion criteria encompassed randomized controlled trials, prospective cohort studies, retrospective observational studies focused adult populations undergoing surgery.Results: The final synthesis included 30 studies. significant reductions in both radicular pain, as well meaningful functional improvements, measured by standardized scales such Neck Disability Index visual analogue scale. Patient satisfaction rates consistently high, most reporting exceeding 85%. Complication low, primarily involving transient neurological deficits typically resolved without need further intervention. Nonrandomized generally presented a moderate risk due confounding selection, whereas trials exhibited low bias.Conclusion: safe effective surgical option disorders substantial pain relief, improvement high levels satisfaction.

Language: Английский

---

Artificial Intelligence-Assisted MRI Diagnosis in Lumbar Degenerative Disc Disease: A Systematic Review

Global Spine Journal, Journal Year: 2024, Volume and Issue: unknown

Published: Aug. 15, 2024

Study Design Systematic review. Objectives Lumbar degenerative disc disease (DDD) poses a significant global health care challenge, with accurate diagnosis being difficult using conventional methods. Artificial intelligence (AI), particularly machine learning and deep learning, offers promising tools for improving diagnostic accuracy workflow in lumbar DDD. This study aims to review AI-assisted magnetic resonance imaging (MRI) DDD discuss current research clinical use. Methods A systematic search of electronic databases identified studies on AI applications MRI-based diagnosis, following Preferred Reporting Items reviews Meta-Analyses (PRISMA) guidelines. Search terms included combinations “Artificial Intelligence,” “Machine Learning,” “Deep “Low Back Pain,” “Lumbar,” “Disc,” “Degeneration,” “MRI,” targeting English from January 1, 2010, 2024. Inclusion criteria encompassed experimental observational peer-reviewed journals. Data extraction focused characteristics, techniques, performance metrics, outcomes, quality assessed predefined criteria. Results Twenty met the inclusion criteria, employing various methodologies, including diagnose manifestations such as degeneration, herniation, bulging. models consistently outperformed methods accuracy, sensitivity, specificity, metrics ranging 71.5% 99% across different objectives. Conclusion The algorithm model provides structured framework integrating into routine practice, enhancing precision patient outcomes management. Further validation are needed refine algorithms real-world application diagnosis.

Language: Английский

---

Artificial Intelligence Detection of Cervical Spine Fractures Using Convolutional Neural Network Models

Neurospine, Journal Year: 2024, Volume and Issue: 21(3), P. 833 - 841

Published: Sept. 27, 2024

To develop and evaluate a technique using convolutional neural networks (CNNs) for the computer-assisted diagnosis of cervical spine fractures from radiographic x-ray images. By leveraging deep learning techniques, study might potentially lead to improved patient outcomes clinical decision-making.

Language: Английский

---

Advancing the future of endoscopic spine surgery

Asian Spine Journal, Journal Year: 2025, Volume and Issue: 19(2), P. IX - X

Published: April 29, 2025

Language: Английский

---

Current Trends and Future Directions in Lumbar Spine Surgery: A Review of Emerging Techniques and Evolving Management Paradigms

Journal of Clinical Medicine, Journal Year: 2025, Volume and Issue: 14(10), P. 3390 - 3390

Published: May 13, 2025

Background/Objectives: Lumbar spine surgery has undergone significant technological transformation in recent years, driven by the goals of minimizing invasiveness, improving precision, and enhancing clinical outcomes. Emerging tools—including robotics, augmented reality, computer-assisted navigation, artificial intelligence—have complemented evolution minimally invasive surgical (MIS) approaches, such as endoscopic lateral interbody fusions. Methods: This systematic review evaluates literature from February 2020 to 2025 on procedural innovations LSS. Eligible studies focused degenerative lumbar pathologies, advanced technologies, reported or perioperative Randomized controlled trials, comparative studies, meta-analyses, large case series were included. Results: A total 32 met inclusion criteria. Robotic-assisted demonstrated high accuracy pedicle screw placement (~92–94%) reduced intraoperative blood loss radiation exposure, although long-term outcomes comparable conventional techniques. Intraoperative navigation improved instrumentation while AR enhanced ergonomic workflow surgeon distraction. AI tools showed promise planning, guidance, outcome prediction but lacked definitive evidence superiority. MIS techniques—including discectomy MIS-TLIF—offered loss, shorter hospital stays, faster recovery, with equivalent pain relief, fusion rates, complication profiles compared open procedures. Lateral oblique approaches (XLIF/OLIF) further optimized alignment indirect decompression, favorable metrics. Conclusions: Recent have technical precision efficiency without compromising patient While short-term benefits are clear, advantages cost-effectiveness require investigation. Integration AI, into reflects an ongoing shift toward personalized, data-driven, less care.

Language: Английский

---

Artificial intelligence: a new cutting-edge tool in spine surgery

Asian Spine Journal, Journal Year: 2024, Volume and Issue: 18(3), P. 458 - 471

Published: June 25, 2024

The purpose of this narrative review was to comprehensively elaborate the various components artificial intelligence (AI), their applications in spine surgery, practical concerns, and future directions. Over years, surgery has been continuously transformed aspects, including diagnostic strategies, surgical approaches, procedures, instrumentation, provide better-quality patient care. Surgeons have also augmented expertise with rapidly growing technological advancements. AI is an advancing field that potential revolutionize many aspects surgery. We performed a comprehensive machine learning To on current role literature using PubMed Google Scholar databases for articles published English last 20 years. initial search keywords "artificial intelligence" AND "spine," "machine learning" "deep "spine" extracted total 78, 60, 37 11,500, 4,610, 2,270 Scholar. After screening exclusion unrelated articles, duplicates, non-English 405 were identified. second stage screening, 93 included review. Studies shown can be used analyze data personalized treatment recommendations It provides valuable insights planning surgeries assisting precise maneuvers decisionmaking during procedures. As more become available further advancements, likely improve outcomes.

Language: Английский

---

Artificial Intelligence Classification for Detecting and Grading Lumbar Intervertebral Disc Degeneration

Spine Surgery and Related Research, Journal Year: 2024, Volume and Issue: 8(6), P. 552 - 559

Published: Aug. 5, 2024

Intervertebral disc degeneration (IDD) is a primary cause of chronic back pain and disability, highlighting the need for precise detection grading effective treatment. This study focuses on developing validating convolutional neural network (CNN) with You Only Look Once (YOLO) architecture model using Pfirrmann system to classify grade lumbar intervertebral based magnetic resonance imaging (MRI) scans.

Language: Английский

---

Commentary on “The Utility and Feasibility of Smart Glasses in Spine Surgery: Minimizing Radiation Exposure During Percutaneous Pedicle Screw Insertion”

Neurospine, Journal Year: 2024, Volume and Issue: 21(2), P. 440 - 442

Published: June 27, 2024

Scientific knowledge used to medicine aid in diagnosis, prevention, treatment, and innovation is referred as medical technology.It does this by creating tools, machines, pharmaceuticals using engineering biotechnology methods. 1,2The manufacturing of equipment techniques utilized the field such augmented reality (AR)-assisted real-time visualization spine surgery, neuromonitoring systems, robotics-assisted robotic-assisted pedicle screw placement, intraoperative navigation systems specifically when discussing spinal technology. 1,3-5 Augmented mixed-reality technologies are included smart glasses (SG) for giving surgeons access imaging, guidance, patient information. 6,7 With use these glasses, surgeon may plan navigate surgery more efficiently image on wearable displays closer than those a fluoroscopic monitor, allowing clearer view reducing radiation exposure during percutaneous (PPS) insertion. 7,8This study 9 examines potential usefulness SG surgery.Adoption offers possible way reduce related health concerns, since fluoroscopy-guided treatments increases.The MOVERIO manufactured Epson Co., Ltd.(Tokyo, Japan) series AR devices designed various applications.The latest glass delivers an engaging experience through quality QHD (quad high definition) or 3-dimensional (3D) images.Its binocular lightweight see-through display also keeps you aware your surroundings while viewing content.The objective research, which employed operators with varying degrees experience, was assess how much reduced increased procedural accuracy.Operators alternated between traditional approaches direct insertion PPS into lumbar model bones under supervision, BT-30E COREVISION 3D fluoroscopy system.The non-SG groups' times did not differ significantly, according data.However, especially less experienced operators, considerably decreased duration amount exposure.Additionally, deviation studies showed that impair precision insertion.The introduction addressed critical concerns regarding Neurospine

Language: Английский

---

Herramientas e inteligencias artificiales para la investigación científica.

Dilemas contemporáneos Educación Política y Valores, Journal Year: 2024, Volume and Issue: unknown

Published: Sept. 1, 2024

La investigación científica es una actividad fundamental para los profesores de tiempo completo en las Instituciones Educación Superior (IES), la cual exige considerable inversión revisar artículos, libros y todo tipo fuentes bibliográficas. Este artículo reporta principales herramientas tecnologías inteligencia artificial usadas generación artículos o documentos científicos. Se destacan aquellas que son utilizadas etapa búsqueda información, lectura comprensión documentos, escritura revisión así como traductores revisores gramaticales. Además, se discuten ventajas desventajas ofrecen dichas herramientas, discute eficiencia pueden proporcionar a investigadores.

---