Three-Dimensional Lumbosacral Reconstruction by An Artificial Intelligence-Based Automated MR Image Segmentation for Selecting the Approach of Percutaneous Endoscopic Lumbar Discectomy

BACKGROUND

Assessing the 3-dimensional (3D) relationship between critical anatomical structures and the surgical channel can help select percutaneous endoscopic lumbar discectomy (PELD) approaches, especially at the L5/S1 level. However, previous evaluation methods for PELD were mainly assessed using 2-dimensional (2D) medical images, making the understanding of the 3D relationship of lumbosacral structures difficult. Artificial intelligence based on automated magnetic resonance (MR) image segmentation has the benefit of 3D reconstruction of medical images.

OBJECTIVES

We developed and validated an artificial intelligence-based MR image segmentation method for constructing a 3D model of lumbosacral structures for selecting the appropriate approach of percutaneous endoscopic lumbar discectomy at the L5/S1 level.

STUDY DESIGN

Three-dimensional reconstruction study using artificial intelligence based on MR image segmentation.

SETTING

Spine and radiology center of a university hospital.

METHODS

Fifty MR data samples were used to develop an artificial intelligence algorithm for automatic segmentation. Manual segmentation and labeling of vertebrae bone (L5 and S1 vertebrae bone), disc, lumbosacral nerve, iliac bone, and skin at the L5/S1 level by 3 experts were used as ground truth. Five-fold cross-validation was performed, and quantitative segmentation metrics were used to evaluate the performance of artificial intelligence based on the MR image segmentation method. The comparison analysis of quantitative measurements between the artificial intelligence-derived 3D (AI-3D) models and the ground truth-derived 3D (GT-3D) models was used to validate the feasibility of 3D lumbosacral structures reconstruction and preoperative assessment of PELD approaches.

RESULTS

Artificial intelligence-based automated MR image segmentation achieved high mean Dice Scores of 0.921, 0.924, 0.885, 0.808, 0.886, and 0.816 for L5 vertebrae bone, S1 vertebrae bone, disc, lumbosacral nerves, iliac bone, and skin, respectively. There were no significant differences between AI-3D and GT-3D models in quantitative measurements. Comparative analysis of quantitative measures showed a high correlation and consistency.

LIMITATIONS

Our method did not involve vessel segmentation in automated MR image segmentation. Our study's sample size was small, and the findings need to be validated in a prospective study with a large sample size.

CONCLUSION

We developed an artificial intelligence-based automated MR image segmentation method, which effectively segmented lumbosacral structures (e.g., L5 vertebrae bone, S1 vertebrae bone, disc, lumbosacral nerve, iliac bone, and skin) simultaneously on MR images, and could be used to construct a 3D model of lumbosacral structures for choosing an appropriate approach of PELD at the L5/S1 level.

New horizons for respiratory system diagnoses

Throughout the past decades ultrasonography did not prove to be a procedure of choice if regarded as part of the routine bedside examination. The reason was the assumption defining the lungs and the bone structures as impenetrable by ultrasound. Only during the recent several years has the approach to the use of such tool in clinical daily routines changed dramatically to offer so-called point-of-care ultrasonography (POCUS). Both vertical and horizontal artefacts became valuable sources of information about the patient's clinical condition, assisting therefore the medical practitioner in differential diagnosis and monitoring of the patient. What is important is that the information is delivered in real time, and the procedure itself is non-invasive. The next stage marking the progress made in this area of diagnostic imaging is the development of arti-ficial intelligence (AI) based on machine learning algorithms. This article is intended to present the available, innovative solutions of the ultrasound systems, including Smart B-line technology, to ensure automatic identification process, as well as interpretation of B-lines in the given lung area of the examined patient. The article sums up the state of the art in ultrasound artefacts and AI applied in POCUS.

Whole Slide Images in Artificial Intelligence Applications in Digital Pathology: Challenges and Pitfalls

The use of digitized data in pathology research is rapidly increasing. The whole slide image (WSI) is an indispensable part of the visual examination of slides in digital pathology and artificial intelligence applications; therefore, the acquisition of WSI with the highest quality is essential. Unlike the conventional routine of pathology, the digital conversion of tissue slides and the differences in its use pose difficulties for pathologists. We categorized these challenges into three groups: before, during, and after the WSI acquisition. The problems before WSI acquisition are usually related to the quality of the glass slide and reflect all existing problems in the analytical process in pathology laboratories. WSI acquisition problems are dependent on the device used to produce the final image file. They may be related to the parts of the device that create an optical image or the hardware and software that enable digitization. Post-WSI acquisition issues are related to the final image file itself, which is the final form of this data, or the software and hardware that will use this file. Because of the digital nature of the data, most of the difficulties are related to the capabilities of the hardware or software. Being aware of the challenges and pitfalls of using digital pathology and AI will make pathologists' integration to the new technologies easier in their daily practice or research.

A Pilot Study Implementing a Machine Learning Algorithm to Use Artificial Intelligence to Diagnose Spinal Conditions

BACKGROUND

Chronic spinal pain is the most prevalent chronic disease, with chronic persistent spinal pain lasting longer than one-year reported in 25% to 60% of the patients. Health care expenditures have been escalating and the financial impact on the US economy is growing. Among multiple modalities of treatments available, facet joint interventions and epidural interventions are the most common ones, in addition to surgical interventions and numerous other conservative modalities of treatments. Despite these increasing costs in the diagnosis and management, disability continues to increase. Consequently, algorithmic approaches have been described as providing a disciplined approach to the use of spinal interventional techniques in managing spinal pain. This approach includes evaluative, diagnostic, and therapeutic approaches, which avoids unnecessary care, as well as poorly documented practices.  Recently, techniques involving artificial intelligence and machine learning have been demonstrated to contribute to the improved understanding, diagnosis, and management of both acute and chronic disease in line with well-designed algorithmic approach. The use of artificial intelligence and machine-learning techniques for the diagnosis of spinal pain has not been widely investigated or adopted.

OBJECTIVES

To evaluate whether it is possible to use artificial intelligence via machine learning algorithms to analyze specific data points and to predict the most likely diagnosis related to spinal pain.

STUDY DESIGN

This was a prospective, observational pilot study.

SETTING

A single pain management center in the United States.

METHODS

A total of 246 consecutive patients with spinal pain were enrolled. Patients were given an iPad to complete a Google form with 85 specific data points, including demographic information, type of pain, pain score, pain location, pain duration, and functional status scores. The data were then input into a decision tree machine learning software program that attempted to learn which data points were most likely to correspond to the practitioner-assigned diagnosis. These outcomes were then compared with the practitioner-assigned diagnosis in the chart.

RESULTS

The average age of the included patients was 57.4 years (range, 18-91 years). The majority of patients were women and the average pain history was approximately 2 years. The most common practitioner-assigned diagnoses included lumbar radiculopathy and lumbar facet disease/spondylosis. Comparison of the software-predicted diagnosis based on reported symptoms with practitioner-assigned diagnosis revealed that the software was accurate approximately 72% of the time.

LIMITATIONS

Additional studies are needed to expand the data set, confirm the predictive ability of the data set, and determine whether it is broadly applicable across pain practices.

CONCLUSIONS

Software-predicted diagnoses based on the data from patients with spinal pain had an accuracy rate of 72%, suggesting promise for augmented decision making using artificial intelligence in this setting.

The Analysis of Pain Research through the Lens of Artificial Intelligence and Machine Learning

BACKGROUND

Traditional pain assessment methods have significant limitations due to the high variability in patient reported pain scores and perception of pain by different individuals. There is a need for generalized and automatic pain detection and recognition methods. In this paper, state-of-the-art machine learning (ML) and deep learning methods in this field are analyzed as well as pain management techniques.

OBJECTIVE

The objective of the study is to analyze the current use of artificial intelligence (AI) and ML in the analysis and management of pain and to disseminate this knowledge prompting future utilization by medical professionals.

STUDY DESIGN

A narrative review of the literature focusing on the latest algorithms in AI and ML for pain assessment and management.

METHODS

Research studies were collected using a literature search on PubMed, Science Direct and IEEE Xplore between 2018 and 2020.

RESULTS

The results of our assessment resulted in the identification of 47 studies meeting inclusion criteria. Pain assessment was the most studied subject with 11 studies, followed by automated measurements with 10 studies, spinal diagnosis with 8 studies, facial expression with 7 studies, pain assessment in special settings evaluated in 5 studies, 4 studies described treatment algorithms, and 2 studies assessed neonatal pain. These studies varied from simple to highly complex methodology. The majority of the studies suffered from inclusion of a small number of patients and without replication of results. However, considering AI and ML are dynamic and emerging specialties, the results shown here are promising. Consequently, we have described all the available literature in summary formats with commentary. Among the various assessments, facial expression and spinal diagnosis and management appear to be ready for inclusion as we continue to progress.

LIMITATIONS

This review is not a systematic review of ML and AI applications in pain research. This review only provides a general idea of the upcoming techniques but does not provide an authoritative evidence-based conclusive opinion of their clinical application and effectiveness.

CONCLUSION

While a majority of the studies focused on classification tasks, very few studies have explored the diagnosis and management of pain. Usage of ML techniques as support tools for clinicians holds an immense potential in the field of pain management.