1
|
Tumko V, Kim J, Uspenskaia N, Honig S, Abel F, Lebl DR, Hotalen I, Kolisnyk S, Kochnev M, Rusakov A, Mourad R. A neural network model for detection and classification of lumbar spinal stenosis on MRI. EUROPEAN SPINE JOURNAL : OFFICIAL PUBLICATION OF THE EUROPEAN SPINE SOCIETY, THE EUROPEAN SPINAL DEFORMITY SOCIETY, AND THE EUROPEAN SECTION OF THE CERVICAL SPINE RESEARCH SOCIETY 2024; 33:941-948. [PMID: 38150003 DOI: 10.1007/s00586-023-08089-2] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Received: 05/31/2023] [Revised: 10/30/2023] [Accepted: 12/04/2023] [Indexed: 12/28/2023]
Abstract
OBJECTIVES To develop a three-stage convolutional neural network (CNN) approach to segment anatomical structures, classify the presence of lumbar spinal stenosis (LSS) for all 3 stenosis types: central, lateral recess and foraminal and assess its severity on spine MRI and to demonstrate its efficacy as an accurate and consistent diagnostic tool. METHODS The three-stage model was trained on 1635 annotated lumbar spine MRI studies consisting of T2-weighted sagittal and axial planes at each vertebral level. Accuracy of the model was evaluated on an external validation set of 150 MRI studies graded on a scale of absent, mild, moderate or severe by a panel of 7 radiologists. The reference standard for all types was determined by majority voting and in case of disagreement, adjudicated by an external radiologist. The radiologists' diagnoses were then compared to the diagnoses of the model. RESULTS The model showed comparable performance to the radiologist average both in terms of the determination of presence/absence of LSS as well as severity classification, for all 3 stenosis types. In the case of central canal stenosis, the sensitivity, specificity and AUROC of the CNN were (0.971, 0.864, 0.963) for binary (presence/absence) classification compared to the radiologist average of (0.786, 0.899, 0.842). For lateral recess stenosis, the sensitivity, specificity and AUROC of the CNN were (0.853, 0.787, 0.907) compared to the radiologist average of (0.713, 0.898, 805). For foraminal stenosis, the sensitivity, specificity and AUROC of the CNN were (0.942, 0.844, 0.950) compared to the radiologist average of (0.879, 0.877, 0.878). Multi-class severity classifications showed similarly comparable statistics. CONCLUSIONS The CNN showed comparable performance to radiologist subspecialists for the detection and classification of LSS. The integration of neural network models in the detection of LSS could bring higher accuracy, efficiency, consistency, and post-hoc interpretability in diagnostic practices.
Collapse
Affiliation(s)
- Vladislav Tumko
- Remedy Logic, 1177 Avenue of the Americas, 5th Floor, New York, NY, 10036, USA
| | - Jack Kim
- Remedy Logic, 1177 Avenue of the Americas, 5th Floor, New York, NY, 10036, USA.
| | - Natalia Uspenskaia
- Remedy Logic, 1177 Avenue of the Americas, 5th Floor, New York, NY, 10036, USA
| | - Shaun Honig
- Remedy Logic, 1177 Avenue of the Americas, 5th Floor, New York, NY, 10036, USA
| | - Frederik Abel
- Hospital for Special Surgery, 535 East 70th Street, New York, NY, 10021, USA
| | - Darren R Lebl
- Hospital for Special Surgery, 535 East 70th Street, New York, NY, 10021, USA
| | - Irene Hotalen
- Remedy Logic, 1177 Avenue of the Americas, 5th Floor, New York, NY, 10036, USA
| | | | - Mikhail Kochnev
- Remedy Logic, 1177 Avenue of the Americas, 5th Floor, New York, NY, 10036, USA
| | - Andrej Rusakov
- Remedy Logic, 1177 Avenue of the Americas, 5th Floor, New York, NY, 10036, USA
| | - Raphaël Mourad
- University of Toulouse, 118 Rte de Narbonne, 31062, Toulouse, France.
| |
Collapse
|
2
|
Park SJ, Hwang JM, Cho DC, Lee S, Kim CH, Han I, Park DW, Kwon HD, Kim KT. Indirect Decompression Using Oblique Lumbar Interbody Fusion Revision Surgery Following Previous Posterior Decompression: Comparison of Clinical and Radiologic Outcomes Between Direct and Indirect Decompression Revision Surgery. Neurospine 2022; 19:544-554. [PMID: 36203280 PMCID: PMC9537844 DOI: 10.14245/ns.2244242.121] [Citation(s) in RCA: 9] [Impact Index Per Article: 4.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/20/2022] [Accepted: 06/02/2022] [Indexed: 12/14/2022] Open
Abstract
OBJECTIVE This study compared the radiological and clinical outcomes with transforaminal lumbar interbody fusion (TLIF) to evaluate the effect of indirect decompression through oblique lumbar interbody fusion (OLIF) as revision surgery. METHODS We enrolled patients who underwent single-level fusion with revision surgery at the same level as the previous decompression level. We retrospectively reviewed 25 patients who underwent OLIF from 2017 to 2018 and 25 who received TLIF from 2014 to 2018. Radiologic and clinical outcomes were evaluated by cross-sectional area (CSA) of the spinal canal, thickness and area of ligamentum flavum (LF), subsidence, disc height, fusion rate, Oswestry Disability Index (ODI), and visual analogue scale (VAS). RESULTS Compared with OLIF, the thickness and area of the LF after surgery were significantly less in TLIF, and the resulting CSA extension was also significantly higher. However, both groups showed improvement in ODI and VAS after surgery, and there was no difference between the groups. Complications related to the posterior approach in TLIF were 4 cases, and in OLIF, there were 2 cases that underwent additional posterior decompression surgery and 6 cases of transient paresthesia. CONCLUSION Since complications associated with the posterior approach can be avoided, OLIF is a safer and useful minimally invasive surgery. Therefore, appropriate indications are applied, OLIF is a good alternative to TLIF when revision surgery is considered.
Collapse
Affiliation(s)
- Sang-Jin Park
- Department of Neurosurgery, Charmjoeun Spine and Joint Hospital, Daegu, Korea
| | - Jong-Moon Hwang
- Department of Rehabilitation Medicine, School of Medicine, Kyungpook National University, Kyungpook National University Hospital, Daegu, Korea
| | - Dae-Chul Cho
- Department of Neurosurgery, School of Medicine, Kyungpook National University, Kyungpook National University Hospital, Daegu, Korea
| | - Subum Lee
- Department of Neurosurgery, Korea University Anam Hospital, Korea University College of Medicine, Seoul, Korea
| | - Chi Heon Kim
- Department of Neurosurgery, Seoul National University Hospital, Seoul, Korea,Department of Neurosurgery and Medical Device Development, Seoul National University College of Medicine, Seoul, Korea
| | - Inbo Han
- Department of Neurosurgery, CHA Bundang Medical Center, CHA University, Seongnam, Korea
| | - Dae-Won Park
- Department of Neurosurgery, Good Moonhwa Hospital, Busan, Korea
| | - Heum-Dai Kwon
- Department of Neurosurgery, Pohang Stroke and Spine Hospital, Pohang, Korea,Co-corresponding Author: Heum-Dai Kwon Department of Neurosurgery, Pohang Stroke and Spine Hospital, 352, Huimangdaero, Nam-gu, Pohang 37659, Korea ,
| | - Kyoung-Tae Kim
- Department of Neurosurgery, School of Medicine, Kyungpook National University, Kyungpook National University Hospital, Daegu, Korea,Corresponding Author Kyoung-Tae Kim Department of Neurosurgery, School of Medicine, Kyungpook National University, Kyungpook National University Hospital, 130 Dongdeok-ro, Jung-gu, Daegu 41944, Korea ,
| |
Collapse
|
3
|
Fushimi Y, Otani K, Tominaga R, Nakamura M, Sekiguchi M, Konno SI. The association between clinical symptoms of lumbar spinal stenosis and MRI axial imaging findings. Fukushima J Med Sci 2021; 67:150-160. [PMID: 34897162 PMCID: PMC8784199 DOI: 10.5387/fms.2021-22] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/30/2022] Open
Abstract
Purpose: In diagnosing lumbar spinal stenosis (LSS), Magnetic Resonance Imaging (MRI) is appropriate to confirm the presence of anatomical stenosis of the spinal canal or compression of the nerve roots. However, it is known that morphological LSS is often present in asymptomatic subjects. There is still controversy about the relationship between anatomical LSS and symptomatic LSS. The aim of this study was to assess the association between qualitative imaging findings on MRI of the lumbar spine and symptomatic LSS. Patients and methods: This was a cross-sectional study of 239 volunteers from an epidemiological survey that included 1,862 participants in total. MRI of the lumbar spine was evaluated in four categories: morphological grading of central stenosis and lateral recess stenosis, presence of the sedimentation sign, and severity of facet joint effusion. The relationship between these morphological evaluations and typical LSS symptoms as assessed by the self-administered, self-reported history questionnaire for lumbar spinal stenosis (LSS-SSHQ) was investigated by multiple logistic regression analysis. Results: The odds ratio of the most severe central stenosis to no stenosis was 15.5 (95%CI: 1.4-164.9). Only the most severe central stenosis was associated with typical LSS symptoms, but not all cases with typical LSS symptoms were due to severe central stenosis. Conclusion: Extreme severe central stenosis was strongly related to typical LSS symptoms. However, although subjects with severe central stenosis showed symptoms suggestive of LSS, these subjects did not always show typical LSS symptoms.
Collapse
Affiliation(s)
- Yuki Fushimi
- Department of Orthopaedic Surgery, Fukushima Medical University School of Medicine
| | - Koji Otani
- Department of Orthopaedic Surgery, Fukushima Medical University School of Medicine
| | - Ryoji Tominaga
- Department of Orthopaedic Surgery, Fukushima Medical University School of Medicine
| | - Masataka Nakamura
- Department of Orthopaedic Surgery, Fukushima Medical University School of Medicine
| | - Miho Sekiguchi
- Department of Orthopaedic Surgery, Fukushima Medical University School of Medicine
| | - Shin-Ichi Konno
- Department of Orthopaedic Surgery, Fukushima Medical University School of Medicine
| |
Collapse
|
4
|
Nikitin AS, Kamchatnov PR. [Neurogenic intermittent claudication in patients with degenerative lumbar stenosis]. Zh Nevrol Psikhiatr Im S S Korsakova 2021; 121:7-11. [PMID: 34184471 DOI: 10.17116/jnevro20211210517] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/17/2022]
Abstract
OBJECTIVE To study the clinical manifestations of neurogenic intermittent claudication (NIC) in patients with degenerative lumbar stenosis (DLS). MATERIAL AND METHODS The neurological status, intensity and nature of pain syndrome, and pain-free walking distance were evaluated in 83 NIC patients with DPS, who underwent surgical treatment. RESULTS The predominance of the clinical picture of sensorimotor pain (n=30; 36.1%) and isolated pain (n=31; 37.3%) forms of NIC was established. Isolated sensory (n=1; 1.2%), sensorimotor (n=3; 3.6%), and motor-pain (n=4; 4.8%) forms were significantly less common. In 6 (7.2%) patients, there was a pain-free form of NIC. In most patients, surgical treatment led to regression of neurological symptoms and an increase in the distance of pain-free walking. No relationship was found between the demographic and clinical characteristics of patients (gender, age, pain severity, pain-free walking distance), MRI results and the features of NIC clinical picture. In 4 patients, the phenomenon of pacing was revealed in the form of an increase in pain at the beginning of walking and a significant decrease or complete regression as walking continues. CONCLUSION The clinical picture of NIC is polymorphic, its pain-free course is possible. The decision on the feasibility of surgical treatment of patients with PLS is possible taking into account the clinical picture of NIC and MRI results.
Collapse
Affiliation(s)
- A S Nikitin
- Evdokimov Moscow State University of Medicine and Dentistry, Moscow, Russia
| | - P R Kamchatnov
- Pirogov Russian National Research Medical University Moscow, Russi, Pirogov Russian National Research Medical University Moscow, Russia
| |
Collapse
|
5
|
Usefulness of Oblique Lumbar Interbody Fusion as Revision Surgery: Comparison of Clinical and Radiological Outcomes Between Primary and Revision Surgery. World Neurosurg 2021; 149:e1067-e1076. [PMID: 33444834 DOI: 10.1016/j.wneu.2020.12.172] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.7] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/30/2020] [Revised: 12/29/2020] [Accepted: 12/30/2020] [Indexed: 11/22/2022]
Abstract
OBJECTIVE Oblique lumbar interbody fusion (OLIF) is useful as surgical treatment of degenerative lumbar disease. However, revision surgery has often resulted in worse surgical outcomes than primary surgery. Thus, we compared the usefulness of OLIF as primary surgery (PS) versus revision surgery (RS). METHODS We retrospectively investigated 173 patients who had undergone single-level OLIF from 2016 to 2018. The radiological and clinical outcomes were compared between PS (n = 152) and RS (n = 21). The effects of RS on the clinical outcomes (Oswestry Disability Index [ODI] cutoff, 12) after surgery were investigated. RESULTS The ODI and visual analog scale score at 6 and 12 months after surgery was worse in the RS group than in the PS group (P < 0.05). In the RS group, the visual analog scale score for leg pain of the previous laminectomy side was worse than that of the virgin side at 6 and 12 months after surgery (P < 0.05). The disc height, ligamentum flavum, and subsidence did not differ between the 2 groups. However, the cross-sectional area enlargement differed between the 2 groups (P < 0.05). Multivariate logistic regression analysis showed that RS and severe subsidence were risk factors for differences in the ODI (P = 0.006 and P = 0.017, respectively). CONCLUSIONS Most radiological outcomes were similar between the RS and PS groups, with no differences in complications or the requirement for additional posterior decompression. However, OLIF resulted in relatively poor clinical outcomes when used as RS. Thus, revision spine surgery tends to result in poor outcomes compared with those of primary spine surgery; however, OLIF can be a tolerable option for revision spine surgery.
Collapse
|