Neurology India, Journal Year: 2023, Volume and Issue: 71(5), P. 872 - 874
Published: Sept. 1, 2023
Language: Английский
Exploration of Targeted Anti-tumor Therapy, Journal Year: 2023, Volume and Issue: unknown, P. 344 - 354
Published: April 28, 2023
Oncologic emergencies are a wide spectrum of oncologic conditions caused directly by malignancies or their treatment. may be classified according to the underlying physiopathology in metabolic, hematologic, and structural conditions. In latter, radiologists have pivotal role, through an accurate diagnosis useful provide optimal patient care. Structural involve central nervous system, thorax, abdomen, emergency know characteristics imaging findings each one them. The number is growing due increased incidence general population also improved survival these patients thanks advances cancer Artificial intelligence (AI) could solution assist with this rapidly increasing workload. To our knowledge, AI applications setting mostly underexplored, probably relatively low difficulty training algorithms. However, defined cause not specific pattern radiological symptoms signs. Therefore, it can expected that algorithms developed for detection non-oncological field transferred clinical emergency. review, craniocaudal approach was followed thoracic, abdominal been addressed regarding reported literature. Among system emergencies, brain herniation spinal cord compression. thoracic district were pulmonary embolism, cardiac tamponade pneumothorax. Pneumothorax most frequently described application AI, improve sensibility reduce time-to-diagnosis. Finally, hemorrhage, intestinal obstruction, perforation, intussusception described.
Language: Английский
Citations
4
Diagnostics, Journal Year: 2024, Volume and Issue: 14(21), P. 2458 - 2458
Published: Nov. 3, 2024
Objectives: The integration of machine learning and radiomics in medical imaging has significantly advanced diagnostic prognostic capabilities healthcare. This study focuses on developing validating an artificial intelligence (AI) model using U-Net architectures for the accurate detection segmentation spinal metastases from computed tomography (CT) images, addressing both osteolytic osteoblastic lesions. Methods: Our methodology employs multiple variations architecture utilizes two distinct datasets: one consisting 115 polytrauma patients vertebra another comprising 38 with documented lesion detection. Results: demonstrated strong performance segmentation, achieving Dice Similarity Coefficient (DSC) values between 0.87 0.96. For metastasis achieved a DSC 0.71 F-beta score 0.68 lytic lesions but struggled sclerotic lesions, obtaining 0.61 0.57, reflecting challenges detecting dense, subtle bone alterations. Despite these limitations, successfully identified isolated metastatic beyond spine, such as sternum, indicating potential broader skeletal Conclusions: concludes that AI-based models can augment radiologists’ by providing reliable second-opinion tools, though further refinements diverse training data are needed optimal performance, particularly segmentation. annotated CT dataset produced shared this research serves valuable resource future advancements.
Language: Английский
Citations
1
North American Spine Society Journal (NASSJ), Journal Year: 2024, Volume and Issue: 19, P. 100513 - 100513
Published: July 3, 2024
Metastasis to the spinal column is a common complication of malignancy, potentially causing pain and neurologic injury. An automated system identify refer patients with metastases can help overcome barriers timely treatment. We describe training, optimization validation natural language processing algorithm presence vertebral metastasis metastatic epidural cord compression (MECC) from radiology reports MRIs.
Language: Английский
Citations
1
World Neurosurgery, Journal Year: 2023, Volume and Issue: 175, P. e823 - e831
Published: April 13, 2023
Language: Английский
Citations
3
European Spine Journal, Journal Year: 2023, Volume and Issue: 32(11), P. 3815 - 3824
Published: April 24, 2023
Language: Английский
Citations
3
Neuroimaging Clinics of North America, Journal Year: 2023, Volume and Issue: 33(3), P. 443 - 457
Published: May 10, 2023
Language: Английский
Citations
3
Radiologic Clinics of North America, Journal Year: 2023, Volume and Issue: 62(2), P. 355 - 370
Published: Nov. 18, 2023
Language: Английский
Citations
3
Published: Jan. 1, 2024
Language: Английский
Citations
0
Cureus, Journal Year: 2024, Volume and Issue: unknown
Published: Sept. 24, 2024
Metastatic epidural spinal cord compression (MESCC) is a critical oncologic emergency caused by the invasion of metastatic tumors into space, leading to cord. If not promptly diagnosed and treated, MESCC can result in irreversible neurological deficits, including paralysis, significantly impacting patient's quality life. Early detection timely intervention are crucial prevent permanent damage. Imaging modalities play pivotal role diagnosis, assessment disease extent, treatment planning for MESCC. Magnetic resonance imaging (MRI) current gold standard due its superior ability visualize cord, lesions. However, recent advances technologies have enhanced management Innovations such as functional MRI, diffusion-weighted (DWI), hybrid techniques like positron emission tomography-computed tomography (PET-CT) PET-MRI improved accuracy particularly detecting early changes guiding therapeutic interventions. This review provides comprehensive analysis evolution MESCC, focusing on their roles detection, planning. It also discusses impact these clinical outcomes future research directions Understanding advancements optimizing improving patient prognosis.
Language: Английский
Citations
0
Frontiers in Neuroscience, Journal Year: 2024, Volume and Issue: 18
Published: Dec. 6, 2024
The purpose of our study is to develop a deep learning (DL) model based on MRI and analyze its consistency with the treatment recommendations for degenerative cervical spine disorders provided by surgeons at hospital.
Language: Английский
Citations
0