Bartłomiej Błaszczyk

The evolution of artificial intelligence (AI) and large language models (LLMs) is updating medical and neurosurgical education. This narrative review provides an updated synthesis of recent applications of AI in neurosurgical education, with particular emphasis on LLMs, knowledge assessment, simulation-based training, automated competency evaluation, and implementation risks. Through machine learning, deep learning, and virtual reality systems, AI enables personalized, adaptive learning pathways, 24/7 on-demand academic assistance, and objective, data-driven competency evaluations in risk-free environments. In neurosurgery, intelligent surgical simulators and automated 3D spatial planning tools optimize technical execution and track procedural metrics. Generative AI streamlines curriculum design, literature synthesis, and examination drafting. However, the adoption of foundation models introduces critical limitations and risks. These systems remain vulnerable to hallucinations, logic confabulations, and fabricated bibliographic citations, which pose risks in high-stakes clinical contexts. Excessive dependence on AI can cause cognitive fatigue, diminish independent critical thinking, and elevate risks related to academic plagiarism, algorithmic bias, and data privacy. To address these challenges, continuous human verification is required. AI should function as a supportive, transparent assistant of human clinical judgment, ensuring that technological innovation reinforces core values of medical practice and patient safety.

Background/Objectives: Transcranial magnetic stimulation (TMS) initiates a cascade from intracranial electric-field exposure through neural recruitment and plasticity to distributed network responses. Neuronavigation improves the geometric reproducibility of delivery but does not guarantee equivalent cortical exposure or target engagement. This narrative review integrates these levels within an operational framework for precision TMS. Methods: Six domain-specific PubMed searches covering 1 January 1985 to 31 July 2026 were supplemented by Google Scholar and citation tracking. A documented rerun on 17 August 2026 yielded 6430 records (5617 unique after cross-query deduplication). Evidence was synthesized narratively; no quantitative synthesis or formal risk-of-bias assessment was performed. Results: Neuronavigation improves geometric precision by stabilizing target definition and coil pose, whereas individualized electric-field models estimate intracranial exposure. Neither establishes biological precision, which also depends on neuronal orientation, brain state, circuit architecture, medication, and behavior. Motor-system measures are not validated as universal biomarkers for nonmotor cortex, and no single validated biomarker captures TMS-induced plasticity. Convergent, controlled multimodal evidence may strengthen inference about target engagement; adaptive and closed-loop approaches remain experimental. Conclusions: Geometric delivery, modeled exposure, biological engagement, and durable functional or clinical benefit require separate validation. Spatial accuracy alone does not establish clinical value.

Background/Objectives: Repetitive transcranial magnetic stimulation (rTMS) has been increasingly investigated as an adjunctive intervention in neurorehabilitation, particularly for motor recovery after stroke. However, conventional rTMS protocols remain limited by variability in target localization, inter-individual anatomical differences, lesion-related network reorganization, and limited reproducibility across treatment sessions. Neuronavigated repetitive transcranial magnetic stimulation (nrTMS) integrates structural neuroimaging with real-time coil tracking, enabling more precise and reproducible stimulation of patient-specific cortical targets. This approach may be especially relevant in patients with focal brain lesions, postoperative anatomical distortion, or functionally reorganized networks. Methods: This narrative review summarizes the biological rationale, current clinical evidence, practical workflow, and limitations of nrTMS in neurorehabilitation, with particular attention to the distinction between conventional rTMS and neuronavigated protocols. Results: The strongest evidence for rTMS-based rehabilitation remains in post-stroke motor recovery, although most studies have used non-navigated protocols. In contrast, postoperative neuro-oncological rehabilitation represents a clinically relevant but still investigational context for nrTMS, as preoperative functional mapping, postoperative deficits, and early rehabilitation can be integrated within a patient-specific therapeutic pathway. Early studies suggest feasibility when stimulation is combined with structured physiotherapy; however, the available evidence is based on small and heterogeneous cohorts, and clinically meaningful superiority over conventional rTMS or standard rehabilitation has not yet been established. Data in traumatic brain injury, multiple sclerosis, ataxias, and neurodegenerative disorders are still preliminary and heterogeneous. Conclusions: Neuronavigation should not be interpreted as an independent therapeutic breakthrough, but rather as a precision-enhancing component of rTMS-based rehabilitation. Its main potential value lies in improving targeting accuracy, session-to-session reproducibility, and integration with individualized neuroimaging and rehabilitation goals. Accordingly, nrTMS should currently be considered a precision-enhancing and hypothesis-generating framework rather than an established rehabilitation standard.

Wstęp: Guzy mózgu stanowią złożoną grupę nowotworów wywodzących się z różnych typów komórek ośrodkowego układu nerwowego. Ich klasyfikacja, oparta na cechach histopatologicznych i molekularnych, determinuje strategie terapeutyczne oraz rokowanie. Postępy w neuroobrazowaniu i technikach biopsji zwiększyły precyzję diagnostyczną, umożliwiając bardziej spersonalizowane i skuteczne metody leczenia. Biopsja mózgu (brain biopsy – BB) jest nieodzownym zabiegiem neurochirurgicznym, służącym do histologicznej diagnostyki nowotworowych zmian mózgu, odgrywającym kluczową rolę w postępowaniu z pacjentem. Celem badania jest ocena wyników histologicznych BB oraz identyfikacja grup wiekowych, rozkładu płci, topografii i różnych histologicznych typów guzów mózgu. Materiał i metody: Przeprowadzono retrospektywne badanie kohortowe w jednym z akademickich ośrodków medycznych, analizując 112 pacjentów poddanych BB w okresie od stycznia 2017 r. do sierpnia 2023 r. Badanie koncentrowało się na wynikach histologicznych oraz markerach molekularnych uzyskanych z biopsji stereotaktycznych i z użyciem neuronawigacji, oceniając skuteczność w uzyskiwaniu próbek diagnostycznych oraz zastosowanie systemu klasyfikacji nowotworów mózgu opracowanego przez Światową Organizację Zdrowia (World Health Organization – WHO). Analizowano również korelacje pomiędzy wynikami histologicznymi a powikłaniami. Wyniki: Badanie histologiczne potwierdziło wysoką skuteczność diagnostyczną BB, przy zbliżonej skuteczności metod stereotaktycznych i z zastosowaniem neuronawigacji. Do kategoryzacji zmian zastosowano klasyfikację nowotworów WHO, co również pozwoliło na zaplanowanie dalszego leczenia onkologicznego. Zaobserwowano niski odsetek powikłań, co potwierdza bezpieczeństwo procedury. Wnioski: Wyniki wskazują, że spektrum rozpoznań nowotworów mózgu u badanych pacjentów jest zbieżne z globalnymi trendami epidemiologicznymi. Zastosowanie klasyfikacji WHO umożliwiło zwiększenie precyzji diagnostycznej oraz ułatwiło standaryzację decyzji terapeutycznych.

The improvement in peripheral nerve repair is still challenging, with the process being difficult and frequently unsatisfying. Injuries, even minor ones, can lead to limitations, including the loss of important life functions such as fingers, hands, or all limbs. Our previous research on animals revealed that the distal part of autologous predegenerated nerve grafts, which were injured and left in place for 7 days, was more capable of supporting reconstructed nerve regeneration. Little is known about the efficacy of using predegenerated autologous grafts in humans. Encouraged by promising results in animal models, we decided to investigate this process in humans. A total of 31 patients were evaluated in the study; 19 predegenerated (injured and left in situ for 7 days) autologous sural nerve implants and 12 fresh sural nerve implants were used, and the period of 2 years after operation was chosen as the time of final clinical assessment. Clinical assessment and motor and sensory nerve conduction velocity were assessed. All data were statistically analyzed using stepwise regression testing and a one-way analysis of variance followed by Tukey’s test for continuous values and the Mann–Whitney U test for ordinal values. Differences were considered statistically significant for p ≤ 0.05. It turns out that autologous, predegenerated sural nerve grafts used for the reconstruction of traumatic peripheral nerves results in better quantitative and qualitative clinical functional outcomes and more adequate nerve conduction parameters.

Background and Clinical Significance: The term “boomerang sign” refers to a boomerang-shaped area of cytotoxic edema in the splenium of the corpus callosum. It is seen as hyperintense lesions on T2-weighted images, FLAIR and DWI in MRI. No specific pathomechanism leading to these changes in the splenium have been yet found; however, authors have listed a variety of potential causes. Case Presentation: The case presents a 38-year-old male patient after left cerebellopontine angle tumor resection with an abnormal, increased signal intensity within the corpus callosum (boomerang sign) in FLAIR MRI sequence. In the case of our patient, unlike the patients described in the literature, the changes in the commissure persist. Conclusions: These lesions could be caused by several factors such as the development of cerebellar edema and subarachnoid bleeding or hypertonic salt usage while in the intensive care unit.

WPROWADZENIE: W ostatnim czasie obserwuje się wzrost liczby opublikowanych artykułów dotyczących sztucznej inteligencji w dziedzinie medycyny, szczególnie w obszarze neurochirurgii. Badania dotyczące integracji sztucznej inteligencji z praktyką neurochirurgiczną wskazują na postępującą zmianę w kierunku szerszego wykorzystania narzędzi wspomaganych sztuczną inteligencją w diagnostyce, analizie obrazu i podejmowaniu decyzji.MATERIAŁ I METODY: W badaniu oceniono efektywność ChatGPT-3.5 i ChatGPT-4 na Państwowym Egzaminie Specjalizacyjnym (PES) z neurochirurgii przeprowadzonym jesienią 2017 r., który w czasie przeprowadzania badania był najnowszym dostępnym na stronie Centrum Egzaminów Medycznych (CEM) egzaminem z oficjalnie udostępnionymi odpowiedziami. Próg zdawalności egzaminu specjalizacyjnego wynosi 56% poprawnych odpowiedzi. Egzamin składał się ze 116 pytań jednokrotnego wyboru, po wyeliminowaniu czterech z uwagi na ich niezgodność z aktualną wiedzą. Ze względu na poruszane zagadnienia pytania podzielono na dziesięć grup tematycznych. Na potrzeby gromadzenia danych obie wersje ChatGPT zostały poinformowane o zasadach egzaminu i poproszone o ocenę stopnia pewności co do każdej odpowiedzi w skali od 1 (zdecydowanie niepewny) do 5 (zdecydowanie pewny). Wszystkie interakcje odbywały się w języku polskim i były rejestrowane.WYNIKI: ChatGPT-4 wyraźnie przewyższył ChatGPT-3.5 z różnicą wynoszącą 29,4% (p < 0,001). W przeciwieństwie do ChatGPT-3.5, ChatGPT-4 z sukcesem osiągnął próg zdawalności dla PES. W testach ChatGPT-3.5 i ChatGPT-4 odpowiedzi były takie same w 61 pytaniach (52,58%), w obu przypadkach były poprawne w 28 pytaniach (24,14%) i niepoprawne w 33 pytaniach (28,45%).WNIOSKI: ChatGPT-4 osiąga większą poprawność w udzielanych odpowiedziach w porównaniu z ChatGPT-3.5, prawdopodobnie dzięki zaawansowanym algorytmom i szerszemu zbiorowi danych treningowych, co podkreśla lepsze zrozumienie złożonych koncepcji neurochirurgicznych.

Background and Objectives: As brain lesions present complex diagnostic challenges, accurate tissue sampling via biopsy is critical for effective treatment planning. Traditional frame-based stereotactic biopsy has been complemented by navigated biopsy techniques, leveraging advancements in imaging and navigation technology. This study aims to compare the navigated and frame-based stereotactic biopsy methods in a clinical setting, evaluating their efficacy, safety, and diagnostic outcomes to determine the optimal approach for precise brain lesion targeting. Materials and Methods: retrospective analysis was conducted on patients who underwent brain biopsies between January 2017 and August 2023 at an academic medical center. Data on patient demographics, clinical characteristics, biopsy technique (navigated vs. frame-based), and outcomes including accuracy, complications, and hospital stay duration were analyzed. Results: The cohort comprised 112 patients, with no significant age or gender differences between groups. Symptoms leading to biopsy were predominantly diminished muscle strength (42.0%), cognitive issues (28.6%), and aphasia (24.1%). Tumors were most common in the deep hemisphere (24.1%). The median hospital stay was 5 days, with a rehospitalization rate of 27.7%. Complications occurred in 4.47% of patients, showing no significant difference between biopsy methods. However, navigated biopsies resulted in fewer samples (p < 0.001) but with comparable diagnostic accuracy as frame-based biopsies. Conclusions: Navigated and frame-based stereotactic biopsies are both effective and safe, with comparable accuracy and complication rates. The choice of technique should consider lesion specifics, surgeon preference, and technological availability. The findings highlight the importance of advanced neurosurgical techniques in enhancing patient care and outcomes.

Background and Objectives: Brain metastases (BMs) pose significant clinical challenges in systemic cancer patients. They often cause symptoms related to brain compression and are typically managed with multimodal therapies, such as surgery, chemotherapy, whole brain radiotherapy (WBRT), and stereotactic radiosurgery (SRS). With modern oncology treatments prolonging survival, concerns about the neurocognitive side effects of BM treatments are growing. WBRT, though widely used for multiple BMs, has recognized neurocognitive toxicity. SRS, particularly Gamma Knife (GK) therapy, offers a minimally invasive alternative with fewer side effects, suitable for patients with a quantifiable number of metastases and better prognoses. Materials and Methods: A retrospective analysis was conducted on 94 patients with multiple BMs treated exclusively with GK at an academic medical center. Patients with prior WBRT were excluded. This study focused on the mean radiation dose received by the hippocampal area, estimated according to the ‘Hippocampal Contouring: A Contouring Atlas for RTOG 0933’ guidelines. Results: The precision of GK equipment results in mean doses of radiation that are lower than those suggested by RTOG 0933 and observed in other studies. This precision may help mitigate cognitive dysfunction and other side effects of hippocampal irradiation. Conclusions: GK therapy facilitates the administration of smaller, safer radiation doses to the hippocampi, which is advantageous even for lesions in the temporal lobe. It is feasible to treat multiple metastases, including cases with more than 10, but it is typically reserved for patients with fewer metastases, with an average of 3 in this study. This underlines GK’s potential for reducing adverse effects while managing BMs effectively.

Embolization is the preferred method for treating intracranial aneurysms due to its less invasive nature. However, recent findings suggest that even uncomplicated embolization may cause structural damage to the brain through ischemic or inflammatory mechanisms. This study aimed to find possible biomarkers of brain injury and inflammation in patients suffering from intracranial aneurysms who underwent endovascular treatment by measuring serological markers indicating brain damage. The study involved 26 patients who underwent uncomplicated intravascular stenting for unruptured intracranial aneurysms between January 2020 and December 2021. Blood samples were collected before the procedure, at 6–12 h, and at 24 h after the procedure. The following protein biomarkers levels were tested with ELISA: S100B, hNSE, TNF, hsCRP, FABP7, NFL, and GP39. Statistical analysis of the results revealed significant increases in serum levels for the four biomarkers: FABP7—before 0.25 (ng/mL) vs. 6–12 h 0.26 (p = 0.012) and vs. 24 h 0.27 (p < 0.001); GP39—before 0.03 (pg/mL) vs. 6–12 h 0.64 (p = 0.011) and vs. 24 h 0.57 (p = 0.001); hsCRP—before 1.65 (μg/mL) vs. 24 h 4.17 (p = 0.037); NFL—before 0.01 (pg/mL) vs. 6–12 h 3.99 (p = 0.004) and vs. 24 h 1.86 (p = 0.033). These biomarkers are recognized as potential indicators of neurovascular damage and should be monitored in clinical settings. Consequently, serum levels of NFL, GP39, hsCRP, and FABP7 measured before and 24 h after endovascular procedures can serve as important markers for assessing brain damage and indicate avenues for further research on biomarkers of neurovascular injury.