RESUMEN
OBJECTIVES: Evaluate the performance of a deep learning (DL)-based model for multiple sclerosis (MS) lesion segmentation and compare it to other DL and non-DL algorithms. METHODS: This ambispective, multicenter study assessed the performance of a DL-based model for MS lesion segmentation and compared it to alternative DL- and non-DL-based methods. Models were tested on internal (n = 20) and external (n = 18) datasets from Latin America, and on an external dataset from Europe (n = 49). We also examined robustness by rescanning six patients (n = 6) from our MS clinical cohort. Moreover, we studied inter-human annotator agreement and discussed our findings in light of these results. Performance and robustness were assessed using intraclass correlation coefficient (ICC), Dice coefficient (DC), and coefficient of variation (CV). RESULTS: Inter-human ICC ranged from 0.89 to 0.95, while spatial agreement among annotators showed a median DC of 0.63. Using expert manual segmentations as ground truth, our DL model achieved a median DC of 0.73 on the internal, 0.66 on the external, and 0.70 on the challenge datasets. The performance of our DL model exceeded that of the alternative algorithms on all datasets. In the robustness experiment, our DL model also achieved higher DC (ranging from 0.82 to 0.90) and lower CV (ranging from 0.7 to 7.9%) when compared to the alternative methods. CONCLUSION: Our DL-based model outperformed alternative methods for brain MS lesion segmentation. The model also proved to generalize well on unseen data and has a robust performance and low processing times both on real-world and challenge-based data. CLINICAL RELEVANCE STATEMENT: Our DL-based model demonstrated superior performance in accurately segmenting brain MS lesions compared to alternative methods, indicating its potential for clinical application with improved accuracy, robustness, and efficiency. KEY POINTS: ⢠Automated lesion load quantification in MS patients is valuable; however, more accurate methods are still necessary. ⢠A novel deep learning model outperformed alternative MS lesion segmentation methods on multisite datasets. ⢠Deep learning models are particularly suitable for MS lesion segmentation in clinical scenarios.
Asunto(s)
Imagen por Resonancia Magnética , Esclerosis Múltiple , Humanos , Imagen por Resonancia Magnética/métodos , Esclerosis Múltiple/diagnóstico por imagen , Esclerosis Múltiple/patología , Redes Neurales de la Computación , Algoritmos , Encéfalo/diagnóstico por imagen , Encéfalo/patologíaRESUMEN
The pudendal nerve entrapment is an entity understudied by diagnosis imaging. Various causes are recognized in relation to difficult labors, rectal, perineal, urological and gynecological surgery, pelvic trauma fracture, bones tumors and compression by tumors or pelvic pseudotumors. Pudendal neuropathy should be clinically suspected, and confirmed by different methods such as electrofisiological testing: evoked potentials, terminal motor latency test and electromyogram, neuronal block and magnetic resonance imaging. The radiologist should be acquainted with the complex anatomy of the pelvic floor, particularly on the path of pudendal nerve studied by magnetic resonance imaging. High resolution magnetic resonance neurography should be used as a complementary diagnostic study along with clinical and electrophysiological examinations in patients with suspected pudendal nerve neuralgia.
Asunto(s)
Imagen por Resonancia Magnética , Nervio Pudendo/diagnóstico por imagen , Neuralgia del Pudendo/diagnóstico por imagen , Diagnóstico Diferencial , Electromiografía , Humanos , Neuroimagen/métodos , Nervio Pudendo/anatomía & histología , Neuralgia del Pudendo/etiología , Neuralgia del Pudendo/terapiaRESUMEN
La neuralgia del nervio pudendo (NP) es una entidad poco estudiada por imágenes. Se reconocen varias causas, tales como compresión a través de su paso por estructuras ligamentarias; estiramiento por partos laboriosos; lesiones secundarias a cirugías rectales, perineales, urológicas y ginecológicas, traumatismos con o sin fractura de huesos pelvianos; procesos inflamatorios/autoinmunes; tumores del NP, y, compresión/desplazamiento por tumores o seudotumores de pelvis. El diagnóstico de neuralgia del NP se sospecha por la clínica y se confirma por diferentes métodos, tales como las pruebas electrofisiolológicas: potenciales evocados, test de latencia motora terminal y electromiograma, y, a través de bloqueos neurales y resonancia magnética. La neurografía por resonancia magnética de alta resolución, debería ser empleada como estudio diagnóstico complementario junto a la clínica y exámenes electrofisiológicos, en los pacientes con sospecha de neuralgia del NP.
The pudendal nerve entrapment is an entity understudied by diagnosis imaging. Various causes are recognized in relation to difficult labors, rectal, perineal, urological and gynecological surgery, pelvic trauma fracture, bones tumors and compression by tumors or pelvic pseudotumors. Pudendal neuropathy should be clinically suspected, and confirmed by different methods such as electrofisiological testing: evoked potentials, terminal motor latency test and electromyogram, neuronal block and magnetic resonance imaging. The radiologist should be acquainted with the complex anatomy of the pelvic floor, particularly on the path of pudendal nerve studied by magnetic resonance imaging. High resolution magnetic resonance neurography should be used as a complementary diagnostic study along with clinical and electrophysiological examinations in patients with suspected pudendal nerve neuralgia.