Radoslav Georgiev

Introduction: Lumbar spine magnetic resonance imaging (MRI) is a high-volume diagnostic examination, yet increasing caseloads and reporting complexity continue to strain radiology workflows. Emerging artificial intelligence (AI)-assisted reading tools may help streamline interpretation and reduce report turnaround times, but their real-world impact on efficiency remains insufficiently quantified. Objective: To evaluate the impact of an AI-based reading tool on lumbar spine MRI interpretation and reporting time. Methods: We randomly selected 236 lumbar spine MRI examinations performed between 2018 and 2023 in patients aged 18 and older. Cases with prior lumbar surgery or scoliosis were excluded. Digital imaging and communications in medicine (DICOM) data were processed using a commercial deep-learning software package, and outputs were reviewed in a standard DICOM viewer. Five radiologists participated. Studies 1 and 2 assessed the effect of AI on interpretation time using a within-reader design: radiologists interpreted each examination with AI support and then reinterpreted the same examinations 2 months later without AI, enabling direct comparison of interpretation times. Study 3 evaluated the effect of AI by comparing AI-assisted and unassisted interpretations in 146 randomly selected examinations. Results: AI assistance significantly accelerated report generation. Across the full dataset, AI-supported interpretation reduced time by approximately 52% compared with unassisted reading. AI-assisted generation of preliminary reports reduced radiologists’ overall time by nearly 30%. Linear mixed-effects modeling indicated that these reductions were statistically significant. The smaller reduction observed in Study 3 (9.21%) may reflect limited familiarity with the software’s reporting style and occasional instances in which the AI outputs did not fully support the radiologists’ findings. Conclusion: AI assistance improves the efficiency of lumbar spine MRI reporting and shortens reporting time.

Pleomorphic xanthoastrocytoma (PXA) is a rare central nervous system malignant neoplasm with a relatively favorable prognosis. As PXA histologically presents with large, multinucleated neoplastic cells, its principal differential diagnosis is giant cell glioblastoma (GCGBM). While there is a significant overlap between the two histologically and the neuropathological diagnosis can be challenging, as well as having some overlap neuroradiologically, the patient prognosis differs significantly, with PXA having a more favorable one. Herein we present a case report of a male patient in his thirties diagnosed with GCGBM and presenting again six years later with thickening of the wall of the porencephalic cyst suggestive of disease recurrence. Histopathology revealed neoplastic spindle, small lymphocyte-like, large epithelioid-like, some with foamy cytoplasm, and scattered large multinucleated cells with bizarre nuclei. For the most part, the tumor had a distinct border to the surrounding brain parenchyma, except for a single zone of invasion. As per the depicted morphology, with a lack of pathognomic features of GCGBM, the diagnosis of PXA was defined, and the oncologic committee reevaluated the patient with treatment reinitiation. Based on the close morphological profile of these neoplasias, it is likely that in the case of limited material, multiple PXA cases are diagnosed as GCGBM, resulting in misdiagnosed long survivors.

Dysembryoplastic neuroepithelial tumors (DNTs) are rare neoplastic entries of the central nervous system. Conventionally DNTs are with cortical location and predominantly occur in the temporal lobe associated with epilepsy. Subtentorial DNTs are rare entries. Herein we report a case of a two-year-old female with a DNT located in the cerebellum. The patient presented clinically with new onset gait instability, headaches and strabismus. Neuroradiology revealed a heterogenous, predominantly cystic lesion in the cerebellar vermis and left cerebellar hemisphere, which was interpreted as possible medulloblastoma based on the patient profile. Frozen section neuropathology was more suggestive of a low-grade glial tumor, with conventional histology and immunohistochemistry showing an admixture of glial and neuronal cells - a complex variety of DNT. Due to the histological appearance, differential diagnosis was required with other neuroglial tumors native to the posterior fossa, such as Lhermitte-Duclos disease. There have been several such published case reports, which, although of older patients, present with similar symptoms and neuropathological findings. The complexity of the neuropathological finding in posterior fossa DNTs can lead to future separation of this entry from conventional DNT, as was seen in the past with septum pellucidum DNT, now referred to as myxoid glioneuronal tumor.

Diffuse astrocytic gliomas and their most common and aggressive representation, glioblastoma (GBM), which as per the 2021 World Health Organization (WHO) guidelines is an isocitrate dehydrogenase (IDH) wildtype without alteration in histone 3 and has glomeruloid vascular proliferation, tumor necrosis, telomerase reverse transcriptase (TERT) promoter mutation, epidermal growth factor receptor (EGFR) gene amplification, or +7/−10 chromosome copy-number changes, are fast-growing tumors with a dismal patient prognosis. Herein, we present cases of a 63-year-old male who, despite no evidence of tumor growth, developed a 6-cm tumor, histologically verified as GBM, WHO CNS grade 4, within eight months, and a 74-year-old female in whom a 1.5-cm tumor grew to 43 mm within 28 days, once again histologically confirmed as GBM, WHO CNS grade 4. Other studies using previous WHO guidelines and including up to 106 cases have shown that these tumors have a daily growth rate of 1.4% and can double their size in a period varying from two weeks to 49.6 days. These growth rates further underline the need for extensive surgical resection as disease progression is rapid, with studies reporting that resection of more than 85% of the tumor volume determined on neuroradiology improves survival compared to biopsy or limited resection and resection of more than 98% of the tumor volume statistically improves patient survival.

The 2021 World Health Organization (WHO) classification of tumors of the central nervous system (CNS) has introduced significant changes to tumor taxonomy. One of the most significant changes in the isolation of isocitrate dehydrogenase (IDH) mutant forms of glioblastoma multiforme (GBM) into separate entities, as well as no longer allowing for entries to be classified as not otherwise specified (NOS). As a result, this entity now includes only the most aggressive adult-type tumors. As such, established prognostic factors no longer apply, as they now form the criteria of different disease entries or have been established based on a mixed cohort. Herein, we aimed to reclassify glioblastoma cases diagnosed per the 2016 WHO tumors of the CNS classification into the 2021 WHO tumors of the CNS classification and establish a patient survival pattern based on age, gender, tumor location, and size as well as tumor O‐6‐methylguanine‐DNA methyltransferase (MGMT) mutation. A retrospective, non-clinical approach was utilized. Biopsy specimens of adults diagnosed with GBM, WHO grade 4, NOS in the period February 2018-February 2021 were reevaluated. The data regarding the patient's gender and age were withdrawn from the medical documentation. Immunohistochemistry was performed with mouse monoclonal anti-IDH R132H and rabbit polyclonal anti-MGMT. Radiology data on tumor location and size were pulled from the radiology repository. Data were statistically analyzed for significance, using Kaplan-Meier survival analysis, with a 95% confidence interval and p<0.05 defined as significant. A total of 58 cases fit the set criteria, with eight of them (13.7%) harboring an IDH R132H mutation and were hence reclassified as diffuse astrocytoma IDH-mutant, WHO CNS grade 4. The cases that retained their GBM classification included n=28 males and n=22 females, a male to female ratio of 1.27:1, and a mean age of 65.3 years (range 43-86 years). The MGMT mutational status revealed a total of n=17 positive cases (35%), while the remaining cases were negative. No hemispheric predilection could be established. Lobar predilection was as follows: temporal (37.78%), parietal (28.89%), frontal (24.44%), and occipital (8.89%). The mean tumor size measured on neuroradiology across the cohort was 50.51 mm (range 20-76 mm). The median survival across cases was 255.96 days (8.41 months), with a range of 18-1150 days (0.59-37.78 months). No statistical correlation could be established between patient survival and gender, hemispheric location, lobar location, and tumor size. A significant difference in survival was established only when comparing the 41-50 age groups to the 71-80 and 81-90 age groups and MGMT positive versus negative tumors (p=0.0001). From a practical standpoint, the changes implemented in the new classification of CNS tumors define GBM as the most aggressive adult type of tumor. Based on their significantly more favorable prognosis, the reclassification of IDH mutant forms of astrocytomas has had little epidemiological impact on this relatively common malignancy but has significantly underlined the dismal prognosis. The changes have also led to MGMT promoter methylation status being the only significant prognostic factor for patient survival in clinical use, based on its prediction for response to temozolomide therapy in this nosological unit clinically presenting when it has already reached immense size.

One of the major components of the functional process in the nasal cavity is taken up by the respiratory epithelium covering the posterior two-thirds of the nasal cavity. Disruption in the cytoarchitectonics and subcellular changes in this epithelium results in a number of functional changes in the nasal cavity. One of the rare and usually iatrogenic disturbances of this type is described in 1996, although noticed and discussed significantly earlier, by Kern and Stenkvist empty nose syndrome (ENS) or secondary atrophic rhinitis. The clinical hallmarks of ENS are described as paradoxical feeling for nasal obstruction in the presence of actually widened nasal airways. This phenomenon is attributed to the efferent neuronal signal dissociation accompanying the changes in the nasal mucosa. Herein we report the findings in a 50-year-old male. The patient presented with chronic right-sided headache, foul discharge and complaints of a stuffed nose in 2011. Endoscopy and computed tomography (CT) showed complete destruction of the hard plane, nasal septum, and right maxillary septum, leading to a formation of a huge neocavity. Due to the past medical history and the severity of the case biopsy specimens were obtained under general anesthesia. The sections showed severe but unspecific changes of the nasal epithelium with areas of minimal remaining preserved respiratory epithelium. Based on the clinical data and endoscopic, CT and histomorphologic data, despite the case is not applicable to the current classification of ENS, the diagnosis of ENS was accepted based on the combined extensive but unspecific findings. A seven-year follow-up period included multiple hospital admissions for monitoring of the condition and extensive sinus lavage. No advancement was noticed.