Mapping the clinical pathway for patients undergoing vestibular schwannoma resection.

BACKGROUND: The introduction of electronic health records (EHR) has improved the collection and storage of patient information, enhancing clinical communication and academia. However, EHRs remain limited by data quality and the time-consuming task of manual data extraction. This study aims to utilise process mapping to help identify critical data entry points within the clinical pathway for VS patients, ideal for structured data entry and automated data collection, in an effort to improve patient care and research. METHODS: A two-stage methodology was conducted at a neurosurgical unit. Process maps were developed using semi-structured interviews with stakeholders in the management of VS resection. Process maps were then retrospectively validated against EHR for patients admitted between August 2019 and December 2021, establishing critical data entry points. RESULTS: Twenty stakeholders were interviewed in the process map development. Process maps were validated against the EHR of 36 patients admitted for VS resection. Operation notes, surgical inpatient reviews (including ward rounds) and discharge summaries were present for all patients, representing critical data entry points. Areas for documentation improvement were present in the preoperative clinics (30/36, 83.3%), preoperative skull base multidisciplinary team (32/36, 88.9%), postoperative follow-up clinics (32/36, 88.9%), and the postoperative skull base multidisciplinary team meeting (29/36, 80.6%). CONCLUSION: This is a first use of a two-stage methodology for process mapping the clinical pathway for patients undergoing VS resection. Our study identified critical data entry points which can be targeted for structured data entry and for automated data collection tools, positively impacting patient care and research.

Mapping patient education encounters in elective surgery: a cohort study and cross-sectional survey.

OBJECTIVE: Develop a process map of when patients learn about their proposed surgery and what resources patients use to educate themselves. DESIGN: A mixed methods design, combining semistructured stakeholder interviews, quantitative validation using electronic healthcare records (EHR) in a retrospective cohort and a cross-sectional patient survey. SETTING: A single surgical centre in the UK. PARTICIPANTS: Fourteen members of the spinal multidisciplinary team were interviewed to develop the process map.This process map was validated using the EHR of 50 patients undergoing elective spine surgery between January and June 2022. Postprocedure, feedback was gathered from 25 patient surveys to identify which resources they used to learn about their spinal procedure. Patients below the age of 18 or who received emergency surgery were excluded. INTERVENTIONS: Elective spine surgery and patient questionnaires given postoperatively either on the ward or in follow-up clinic. PRIMARY AND SECONDARY OUTCOME MEASURES: The primary outcome was the percentage of the study cohort that was present at encounters on the process map. Key timepoints were defined if >80% of patients were present. The secondary outcome was the percentage of the study cohort that used educational resources listed in the patient questionnaire. RESULTS: There were 342 encounters which occurred across the cohort, with 16 discrete event categories identified. The initial surgical clinic (88%), anaesthetic preoperative assessment (96%) and admission for surgery (100%) were identified as key timepoints. Surveys identified that patients most used verbal information from their surgeon (100%) followed by written information from their surgeon (52%) and the internet (40%) to learn about their surgery. CONCLUSIONS: Process mapping is an effective method of illustrating the patient pathway. The initial surgical clinic, anaesthetic preoperative assessment and surgical admission are key timepoints where patients receive information. This has future implications for guiding patient education interventions to focus at key timepoints.

Early evaluation of a natural language processing tool to improve access to educational resources for surgical patients.

PURPOSE: Accessible patient information sources are vital in educating patients about the benefits and risks of spinal surgery, which is crucial for obtaining informed consent. We aim to assess the effectiveness of a natural language processing (NLP) pipeline in recognizing surgical procedures from clinic letters and linking this with educational resources. METHODS: Retrospective examination of letters from patients seeking surgery for degenerative spinal disease at a single neurosurgical center. We utilized MedCAT, a named entity recognition and linking NLP, integrated into the electronic health record (EHR), which extracts concepts and links them to systematized nomenclature of medicine-clinical terms (SNOMED-CT). Investigators reviewed clinic letters, identifying words or phrases that described or identified operations and recording the SNOMED-CT terms as ground truth. This was compared to SNOMED-CT terms identified by the model, untrained on our dataset. A pipeline linking clinic letters to patient-specific educational resources was established, and precision, recall, and F1 scores were calculated. RESULTS: Across 199 letters the model identified 582 surgical procedures, and the overall pipeline after adding rules a total of 784 procedures (precision = 0.94, recall = 0.86, F1 = 0.91). Across 187 letters with identified SNOMED-CT terms the integrated pipeline linking education resources directly to the EHR was successful in 157 (78%) patients (precision = 0.99, recall = 0.87, F1 = 0.92). CONCLUSIONS: NLP accurately identifies surgical procedures in pre-operative clinic letters within an untrained subspecialty. Performance varies among letter authors and depends on the language used by clinicians. The identified procedures can be linked to patient education resources, potentially improving patients' understanding of surgical procedures.

PitSurgRT: real-time localization of critical anatomical structures in endoscopic pituitary surgery.

PURPOSE: Endoscopic pituitary surgery entails navigating through the nasal cavity and sphenoid sinus to access the sella using an endoscope. This procedure is intricate due to the proximity of crucial anatomical structures (e.g. carotid arteries and optic nerves) to pituitary tumours, and any unintended damage can lead to severe complications including blindness and death. Intraoperative guidance during this surgery could support improved localization of the critical structures leading to reducing the risk of complications. METHODS: A deep learning network PitSurgRT is proposed for real-time localization of critical structures in endoscopic pituitary surgery. The network uses high-resolution net (HRNet) as a backbone with a multi-head for jointly localizing critical anatomical structures while segmenting larger structures simultaneously. Moreover, the trained model is optimized and accelerated by using TensorRT. Finally, the model predictions are shown to neurosurgeons, to test their guidance capabilities. RESULTS: Compared with the state-of-the-art method, our model significantly reduces the mean error in landmark detection of the critical structures from 138.76 to 54.40 pixels in a 1280 × 720-pixel image. Furthermore, the semantic segmentation of the most critical structure, sella, is improved by 4.39% IoU. The inference speed of the accelerated model achieves 298 frames per second with floating-point-16 precision. In the study of 15 neurosurgeons, 88.67% of predictions are considered accurate enough for real-time guidance. CONCLUSION: The results from the quantitative evaluation, real-time acceleration, and neurosurgeon study demonstrate the proposed method is highly promising in providing real-time intraoperative guidance of the critical anatomical structures in endoscopic pituitary surgery.

Can ChatGPT outperform a neurosurgical trainee? A prospective comparative study.

PURPOSE: This study aimed to compare the performance of ChatGPT, a large language model (LLM), with human neurosurgical applicants in a neurosurgical national selection interview, to assess the potential of artificial intelligence (AI) and LLMs in healthcare and provide insights into their integration into the field. METHODS: In a prospective comparative study, a set of neurosurgical national selection-style interview questions were asked to eight human participants and ChatGPT in an online interview. All participants were doctors currently practicing in the UK who had applied for a neurosurgical National Training Number. Interviews were recorded, anonymised, and scored by three neurosurgical consultants with experience as interviewers for national selection. Answers provided by ChatGPT were used as a template for a virtual interview. Interview transcripts were subsequently scored by neurosurgical consultants using criteria utilised in real national selection interviews. Overall interview score and subdomain scores were compared between human participants and ChatGPT. RESULTS: For overall score, ChatGPT fell behind six human competitors and did not achieve a mean score higher than any individuals who achieved training positions. Several factors, including factual inaccuracies and deviations from expected structure and style may have contributed to ChatGPT's underperformance. CONCLUSIONS: LLMs such as ChatGPT have huge potential for integration in healthcare. However, this study emphasises the need for further development to address limitations and challenges. While LLMs have not surpassed human performance yet, collaboration between humans and AI systems holds promise for the future of healthcare.

The IDEAL framework for surgical robotics: development, comparative evaluation and long-term monitoring.

The next generation of surgical robotics is poised to disrupt healthcare systems worldwide, requiring new frameworks for evaluation. However, evaluation during a surgical robot's development is challenging due to their complex evolving nature, potential for wider system disruption and integration with complementary technologies like artificial intelligence. Comparative clinical studies require attention to intervention context, learning curves and standardized outcomes. Long-term monitoring needs to transition toward collaborative, transparent and inclusive consortiums for real-world data collection. Here, the Idea, Development, Exploration, Assessment and Long-term monitoring (IDEAL) Robotics Colloquium proposes recommendations for evaluation during development, comparative study and clinical monitoring of surgical robots-providing practical recommendations for developers, clinicians, patients and healthcare systems. Multiple perspectives are considered, including economics, surgical training, human factors, ethics, patient perspectives and sustainability. Further work is needed on standardized metrics, health economic assessment models and global applicability of recommendations.

Benchtop simulation of the retrosigmoid approach: Validation of a surgical simulator and development of a task-specific outcome measure score.

Background: Neurosurgical training is changing globally. Reduced working hours and training opportunities, increased patient safety expectations, and the impact of COVID-19 have reduced operative exposure. Benchtop simulators enable trainees to develop surgical skills in a controlled environment. We aim to validate a high-fidelity simulator model (RetrosigmoidBox, UpSurgeOn) for the retrosigmoid approach to the cerebellopontine angle (CPA). Methods: Novice and expert Neurosurgeons and Ear, Nose, and Throat surgeons performed a surgical task using the model - identification of the trigeminal nerve. Experts completed a post-task questionnaire examining face and content validity. Construct validity was assessed through scoring of operative videos employing Objective Structured Assessment of Technical Skills (OSATS) and a novel Task-Specific Outcome Measure score. Results: Fifteen novice and five expert participants were recruited. Forty percent of experts agreed or strongly agreed that the brain tissue looked real. Experts unanimously agreed that the RetrosigmoidBox was appropriate for teaching. Statistically significant differences were noted in task performance between novices and experts, demonstrating construct validity. Median total OSATS score was 14/25 (IQR 10-19) for novices and 22/25 (IQR 20-22) for experts (p < 0.05). Median Task-Specific Outcome Measure score was 10/20 (IQR 7-17) for novices compared to 19/20 (IQR 18.5-19.5) for experts (p < 0.05). Conclusion: The RetrosigmoidBox benchtop simulator has a high degree of content and construct validity and moderate face validity. The changing landscape of neurosurgical training mean that simulators are likely to become increasingly important in the delivery of high-quality education. We demonstrate the validity of a Task-Specific Outcome Measure score for performance assessment of a simulated approach to the CPA.

A synthetic model simulator for intracranial aneurysm clipping: validation of the UpSurgeOn AneurysmBox.

Background and objectives: In recent decades, the rise of endovascular management of aneurysms has led to a significant decline in operative training for surgical aneurysm clipping. Simulation has the potential to bridge this gap and benchtop synthetic simulators aim to combine the best of both anatomical realism and haptic feedback. The aim of this study was to validate a synthetic benchtop simulator for aneurysm clipping (AneurysmBox, UpSurgeOn). Methods: Expert and novice surgeons from multiple neurosurgical centres were asked to clip a terminal internal carotid artery aneurysm using the AneurysmBox. Face and content validity were evaluated using Likert scales by asking experts to complete a post-task questionnaire. Construct validity was evaluated by comparing expert and novice performance using the modified Objective Structured Assessment of Technical Skills (mOSATS), developing a curriculum-derived assessment of Specific Technical Skills (STS), and measuring the forces exerted using a force-sensitive glove. Results: Ten experts and eighteen novices completed the task. Most experts agreed that the brain looked realistic (8/10), but far fewer agreed that the brain felt realistic (2/10). Half the expert participants (5/10) agreed that the aneurysm clip application task was realistic. When compared to novices, experts had a significantly higher median mOSATS (27 vs. 14.5; p < 0.01) and STS score (18 vs. 9; p < 0.01); the STS score was strongly correlated with the previously validated mOSATS score (p < 0.01). Overall, there was a trend towards experts exerting a lower median force than novices, however, these differences were not statistically significant (3.8 N vs. 4.0 N; p = 0.77). Suggested improvements for the model included reduced stiffness and the addition of cerebrospinal fluid (CSF) and arachnoid mater. Conclusion: At present, the AneurysmBox has equivocal face and content validity, and future versions may benefit from materials that allow for improved haptic feedback. Nonetheless, it has good construct validity, suggesting it is a promising adjunct to training.

Optimising trajectory planning for stereotactic brain tumour biopsy using artificial intelligence: a systematic review of the literature.

PURPOSE: Despite advances in technology, stereotactic brain tumour biopsy remains challenging due to the risk of injury to critical structures. Indeed, choosing the correct trajectory remains essential to patient safety. Artificial intelligence can be used to perform automated trajectory planning. We present a systematic review of automated trajectory planning algorithms for stereotactic brain tumour biopsies. METHODS: A PRISMA adherent systematic review was conducted. Databases were searched using keyword combinations of 'artificial intelligence', 'trajectory planning' and 'brain tumours'. Studies reporting applications of artificial intelligence (AI) to trajectory planning for brain tumour biopsy were included. RESULTS: All eight studies were in the earliest stage of the IDEAL-D development framework. Trajectory plans were compared through a variety of surrogate markers of safety, of which the minimum distance to blood vessels was the most common. Five studies compared manual to automated planning strategies and favoured automation in all cases. However, this comes with a significant risk of bias. CONCLUSIONS: This systematic review reveals the need for IDEAL-D Stage 1 research into automated trajectory planning for brain tumour biopsy. Future studies should establish the congruence between expected risk of algorithms and the ground truth through comparisons to real world outcomes.

Shifted-windows transformers for the detection of cerebral aneurysms in microsurgery.

PURPOSE: Microsurgical Aneurysm Clipping Surgery (MACS) carries a high risk for intraoperative aneurysm rupture. Automated recognition of instances when the aneurysm is exposed in the surgical video would be a valuable reference point for neuronavigation, indicating phase transitioning and more importantly designating moments of high risk for rupture. This article introduces the MACS dataset containing 16 surgical videos with frame-level expert annotations and proposes a learning methodology for surgical scene understanding identifying video frames with the aneurysm present in the operating microscope's field-of-view. METHODS: Despite the dataset imbalance (80% no presence, 20% presence) and developed without explicit annotations, we demonstrate the applicability of Transformer-based deep learning architectures (MACSSwin-T, vidMACSSwin-T) to detect the aneurysm and classify MACS frames accordingly. We evaluate the proposed models in multiple-fold cross-validation experiments with independent sets and in an unseen set of 15 images against 10 human experts (neurosurgeons). RESULTS: Average (across folds) accuracy of 80.8% (range 78.5-82.4%) and 87.1% (range 85.1-91.3%) is obtained for the image- and video-level approach, respectively, demonstrating that the models effectively learn the classification task. Qualitative evaluation of the models' class activation maps shows these to be localized on the aneurysm's actual location. Depending on the decision threshold, MACSWin-T achieves 66.7-86.7% accuracy in the unseen images, compared to 82% of human raters, with moderate to strong correlation. CONCLUSIONS: Proposed architectures show robust performance and with an adjusted threshold promoting detection of the underrepresented (aneurysm presence) class, comparable to human expert accuracy. Our work represents the first step towards landmark detection in MACS with the aim to inform surgical teams to attend to high-risk moments, taking precautionary measures to avoid rupturing.

Process analysis of the patient pathway for automated data collection: an exemplar using pituitary surgery.

Introduction: Automation of routine clinical data shows promise in relieving health systems of the burden associated with manual data collection. Identifying consistent points of documentation in the electronic health record (EHR) provides salient targets to improve data entry quality. Using our pituitary surgery service as an exemplar, we aimed to demonstrate how process mapping can be used to identify reliable areas of documentation in the patient pathway to target structured data entry interventions. Materials and methods: This mixed methods study was conducted in the largest pituitary centre in the UK. Purposive snowball sampling identified frontline stakeholders for process mapping to produce a patient pathway. The final patient pathway was subsequently validated against a real-world dataset of 50 patients who underwent surgery for pituitary adenoma. Events were categorized by frequency and mapped to the patient pathway to determine critical data points. Results: Eighteen stakeholders encompassing all members of the multidisciplinary team (MDT) were consulted for process mapping. The commonest events recorded were neurosurgical ward round entries (N = 212, 14.7%), pituitary clinical nurse specialist (CNS) ward round entries (N = 88, 6.12%) and pituitary MDT treatment decisions (N = 88, 6.12%) representing critical data points. Operation notes and neurosurgical ward round entries were present for every patient. 43/44 (97.7%) had a pre-operative pituitary MDT entry, pre-operative clinic letter, a post-operative clinic letter, an admission clerking entry, a discharge summary, and a post-operative histopathology pituitary multidisciplinary (MDT) team entries. Conclusion: This is the first study to produce a validated patient pathway of patients undergoing pituitary surgery, serving as a comparison to optimise this patient pathway. We have identified salient targets for structured data entry interventions, including mandatory datapoints seen in every admission and have also identified areas to improve documentation adherence, both of which support movement towards automation.

Characterization of patients with idiopathic normal pressure hydrocephalus using natural language processing within an electronic healthcare record system.

OBJECTIVE: Idiopathic normal pressure hydrocephalus (iNPH) is an underdiagnosed, progressive, and disabling condition. Early treatment is associated with better outcomes and improved quality of life. In this paper, the authors aimed to identify features associated with patients with iNPH using natural language processing (NLP) to characterize this cohort, with the intention to later target the development of artificial intelligence-driven tools for early detection. METHODS: The electronic health records of patients with shunt-responsive iNPH were retrospectively reviewed using an NLP algorithm. Participants were selected from a prospectively maintained single-center database of patients undergoing CSF diversion for probable iNPH (March 2008-July 2020). Analysis was conducted on preoperative health records including clinic letters, referrals, and radiology reports accessed through CogStack. Clinical features were extracted from these records as SNOMED CT (Systematized Nomenclature of Medicine Clinical Terms) concepts using a named entity recognition machine learning model. In the first phase, a base model was generated using unsupervised training on 1 million electronic health records and supervised training with 500 double-annotated documents. The model was fine-tuned to improve accuracy using 300 records from patients with iNPH double annotated by two blinded assessors. Thematic analysis of the concepts identified by the machine learning algorithm was performed, and the frequency and timing of terms were analyzed to describe this patient group. RESULTS: In total, 293 eligible patients responsive to CSF diversion were identified. The median age at CSF diversion was 75 years, with a male predominance (69% male). The algorithm performed with a high degree of precision and recall (F1 score 0.92). Thematic analysis revealed the most frequently documented symptoms related to mobility, cognitive impairment, and falls or balance. The most frequent comorbidities were related to cardiovascular and hematological problems. CONCLUSIONS: This model demonstrates accurate, automated recognition of iNPH features from medical records. Opportunities for translation include detecting patients with undiagnosed iNPH from primary care records, with the aim to ultimately improve outcomes for these patients through artificial intelligence-driven early detection of iNPH and prompt treatment.

Neurosurgical Team Acceptability of Brain-Computer Interfaces: A Two-Stage International Cross-Sectional Survey.

OBJECTIVE: Invasive brain-computer interfaces (BCIs) require neurosurgical implantation, which confers a range of risks. Despite this situation, no studies have assessed the acceptability of invasive BCIs among the neurosurgical team. This study aims to establish baseline knowledge of BCIs within the neurosurgical team and identify attitudes toward different applications of invasive BCI. METHODS: A 2-stage cross-sectional international survey of the neurosurgical team (neurosurgeons, anesthetists, and operating room nurses) was conducted. Results from the first, qualitative, survey were used to guide the second-stage quantitative survey, which assessed acceptability of invasive BCI applications. Five-part Likert scales were used to collect quantitative data. Surveys were distributed internationally via social media and collaborators. RESULTS: A total of 108 qualitative responses were collected. Themes included the promise of BCIs positively affecting disease targets, concerns regarding stability, and an overall positive emotional reaction to BCI technology. The quantitative survey generated 538 responses from 32 countries. Baseline knowledge of BCI technology was poor, with 9% claiming to have a good or expert knowledge of BCIs. Acceptability of invasive BCI for rehabilitative purposes was >80%. Invasive BCI for augmentation in healthy populations divided opinion. CONCLUSIONS: The neurosurgical team's view of the acceptability of invasive BCI was divided across a range of indications. Some applications (e.g., stroke rehabilitation) were viewed as more appropriate than other applications (e.g., augmentation for military use). This range in views highlights the need for stakeholder consultation on acceptable use cases along with regulation and guidance to govern initial BCI implantations if patients are to realize the potential benefits.

Fenretinide cytotoxicity is independent of both constitutive and pharmacologically modulated glutathione levels in pediatric acute lymphoblastic leukemia cells cultured at hypoxia.

Fenretinide (4-HPR) cytotoxicity relative to glutathione levels in pediatric acute lymphoblastic leukemia cell lines cultured at bone marrow level hypoxia (5% O2) is evaluated. 4-HPR cytotoxicity correlated with reactive oxygen species generation (P < 0.001),but not with levels of intracellular glutathione, g-glutamylcysteine synthase, or glutathione peroxidase. Buthionine sulfoximine (BSO)reduced glutathione levels in 10 cell lines (P < 0.001), but 4-HPR þ BSO was markedly synergistic in only 1 of 10 lines. Pretreatment with N-acetylcysteine increased glutathione (P < 0.02)but did not alter 4-HPR cytotoxicity. Our data suggest that 4-HPR cytotoxicity is independent of glutathione under physiologic oxygen tension.

Fenretinide metabolism in humans and mice: utilizing pharmacological modulation of its metabolic pathway to increase systemic exposure.

BACKGROUND AND PURPOSE: High plasma levels of fenretinide [N-(4-hydroxyphenyl)retinamide (4-HPR)] were associated with improved outcome in a phase II clinical trial. Low bioavailability of 4-HPR has been limiting its therapeutic applications. This study characterized metabolism of 4-HPR in humans and mice, and to explore the effects of ketoconazole, an inhibitor of CYP3A4, as a modulator to increase 4-HPR plasma levels in mice and to increase the low bioavailability of 4-HPR. EXPERIMENTAL APPROACH: 4-HPR metabolites were identified by mass spectrometric analysis and levels of 4-HPR and its metabolites [N-(4-methoxyphenyl)retinamide (4-MPR) and 4-oxo-N-(4-hydroxyphenyl)retinamide (4-oxo-4-HPR)] were quantified by high-performance liquid chromatography (HPLC). Kinetic analysis of enzyme activities and the effects of enzyme inhibitors were performed in pooled human and pooled mouse liver microsomes, and in human cytochrome P450 (CYP) 3A4 isoenzyme microsomes. In vivo metabolism of 4-HPR was inhibited in mice. KEY RESULTS: Six 4-HPR metabolites were identified in the plasma of patients and mice. 4-HPR was oxidized to 4-oxo-4-HPR, at least in part via human CYP3A4. The CYP3A4 inhibitor ketoconazole significantly reduced 4-oxo-4-HPR formation in both human and mouse liver microsomes. In two strains of mice, co-administration of ketoconazole with 4-HPR in vivo significantly increased 4-HPR plasma concentrations by > twofold over 4-HPR alone and also increased 4-oxo-4-HPR levels. CONCLUSIONS AND IMPLICATIONS: Mice may serve as an in vivo model of human 4-HPR pharmacokinetics. In vivo data suggest that the co-administration of ketoconazole at normal clinical doses with 4-HPR may increase systemic exposure to 4-HPR in humans.

Human transcriptome analysis reveals a potential role for active transport in the metabolism of Pseudomonas aeruginosa autoinducers.

The opportunistic pathogen Pseudomonas aeruginosa employs acyl homoserine lactones (AHL) as signaling compounds to regulate virulence gene expression via quorum sensing. The AHL N-3-oxo-dodecanoyl-l-homoserine lactone (3OC(12)-HSL) also induces mammalian cell responses, including apoptosis and immune modulation. In certain cell types the apoptotic effects of 3OC(12)-HSL are mediated via a calcium-dependent signaling pathway, while some pro-inflammatory effects involve intracellular transcriptional regulators. However, the mechanisms by which mammalian cells perceive and respond to 3OC(12)-HSL are still not completely understood. Here we used microarray analysis to investigate the transcriptional response of human lung epithelial cells after exposure to 3OC(12)-HSL. These data revealed that mRNA levels for several genes involved in xenobiotic sensing and drug transport were increased in cells exposed to 3OC(12)-HSL, which led us to examine the intracellular fate of 3OC(12)-HSL. Using radiolabeled autoinducer uptake assays, we discovered that intracellular 3OC(12)-HSL levels increased after exposure and achieved maximal levels after 20-30 min. Intracellular 3OC(12)-HSL decreased to background levels over the next 90 min and this process was blocked by pre-treatment with an inhibitor of the ABC transporter ABCA1. Taken together, these data suggest that mammalian cells detect 3OC(12)-HSL and activate protective mechanisms to expel it from the cell.