Search

We present an ensemble transfer learning method to predict suicide from Veterans Affairs (VA) electronic medical records (EMR). A diverse set of base models was trained to predict a binary outcome constructed from reported suicide, suicide attempt, and overdose diagnoses with varying choices of study design and prediction methodology. Each model used twenty cross-sectional and 190 longitudinal variables observed in eight time intervals covering 7.5 years prior to the time of prediction. Ensembles of seven base models were created and fine-tuned with ten variables expected to change with study design and outcome definition in order to predict suicide and combined outcome in a prospective cohort. The ensemble models achieved c-statistics of 0.73 on 2-year suicide risk and 0.83 on the combined outcome when predicting on a prospective cohort of [Formula: see text] 4.2 M veterans. The ensembles rely on nonlinear base models trained using a matched retrospective nested case-control (Rcc) study cohort and show good calibration across a diversity of subgroups, including risk strata, age, sex, race, and level of healthcare utilization. In addition, a linear Rcc base model provided a rich set of biological predictors, including indicators of suicide, substance use disorder, mental health diagnoses and treatments, hypoxia and vascular damage, and demographics., (© 2024. This is a U.S. Government work and not under copyright protection in the US; foreign copyright protection may apply.)

Objective: To determine whether automated, electronic alerts increased referrals for epilepsy surgery., Methods: We conducted a prospective, randomized controlled trial of a natural language processing-based clinical decision support system embedded in the electronic health record (EHR) at 14 pediatric neurology outpatient clinic sites. Children with epilepsy and at least two prior neurology visits were screened by the system prior to their scheduled visit. Patients classified as a potential surgical candidate were randomized 2:1 for their provider to receive an alert or standard of care (no alert). The primary outcome was referral for a neurosurgical evaluation. The likelihood of referral was estimated using a Cox proportional hazards regression model., Results: Between April 2017 and April 2019, at total of 4858 children were screened by the system, and 284 (5.8%) were identified as potential surgical candidates. Two hundred four patients received an alert, and 96 patients received standard care. Median follow-up time was 24 months (range: 12-36 months). Compared to the control group, patients whose provider received an alert were more likely to be referred for a presurgical evaluation (3.1% vs 9.8%; adjusted hazard ratio [HR] = 3.21, 95% confidence interval [CI]: 0.95-10.8; one-sided p = .03). Nine patients (4.4%) in the alert group underwent epilepsy surgery, compared to none (0%) in the control group (one-sided p = .03)., Significance: Machine learning-based automated alerts may improve the utilization of referrals for epilepsy surgery evaluations., (© 2023 The Authors. Epilepsia published by Wiley Periodicals LLC on behalf of International League Against Epilepsy.)

Objectives: Epilepsy surgery is underutilized. Automating the identification of potential surgical candidates may facilitate earlier intervention. Our objective was to develop site-specific machine learning (ML) algorithms to identify candidates before they undergo surgery., Materials & Methods: In this multicenter, retrospective, longitudinal cohort study, ML algorithms were trained on n-grams extracted from free-text neurology notes, EEG and MRI reports, visit codes, medications, procedures, laboratories, and demographic information. Site-specific algorithms were developed at two epilepsy centers: one pediatric and one adult. Cases were defined as patients who underwent resective epilepsy surgery, and controls were patients with epilepsy with no history of surgery. The output of the ML algorithms was the estimated likelihood of candidacy for resective epilepsy surgery. Model performance was assessed using 10-fold cross-validation., Results: There were 5880 children (n = 137 had surgery [2.3%]) and 7604 adults with epilepsy (n = 56 had surgery [0.7%]) included in the study. Pediatric surgical patients could be identified 2.0 years (range: 0-8.6 years) before beginning their presurgical evaluation with AUC =0.76 (95% CI: 0.70-0.82) and PR-AUC =0.13 (95% CI: 0.07-0.18). Adult surgical patients could be identified 1.0 year (range: 0-5.4 years) before beginning their presurgical evaluation with AUC =0.85 (95% CI: 0.78-0.93) and PR-AUC =0.31 (95% CI: 0.14-0.48). By the time patients began their presurgical evaluation, the ML algorithms identified pediatric and adult surgical patients with AUC =0.93 and 0.95, respectively. The mean squared error of the predicted probability of surgical candidacy (Brier scores) was 0.018 in pediatrics and 0.006 in adults., Conclusions: Site-specific machine learning algorithms can identify candidates for epilepsy surgery early in the disease course in diverse practice settings., (© 2021 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd.)

Background: As adolescent suicide rates continue to rise, innovation in risk identification is warranted. Machine learning can identify suicidal individuals based on their language samples. This feasibility pilot was conducted to explore this technology's use in adolescent therapy sessions and assess machine learning model performance., Method: Natural language processing machine learning models to identify level of suicide risk using a smartphone app were tested in outpatient therapy sessions. Data collection included language samples, depression and suicidality standardized scale scores, and therapist impression of the client's mental state. Previously developed models were used to predict suicidal risk., Results: 267 interviews were collected from 60 students in eight schools by ten therapists, with 29 students indicating suicide or self-harm risk. During external validation, models were trained on suicidal speech samples collected from two separate studies. We found that support vector machines (AUC: 0.75; 95% CI: 0.69-0.81) and logistic regression (AUC: 0.76; 95% CI: 0.70-0.82) lead to good discriminative ability, with an extreme gradient boosting model performing the best (AUC: 0.78; 95% CI: 0.72-0.84)., Conclusion: Voice collection technology and associated procedures can be integrated into mental health therapists' workflow. Collected language samples could be classified with good discrimination using machine learning methods.

Objective: With early identification and intervention, many suicidal deaths are preventable. Tools that include machine learning methods have been able to identify suicidal language. This paper examines the persistence of this suicidal language up to 30 days after discharge from care., Method: In a multi-center study, 253 subjects were enrolled into either suicidal or control cohorts. Their responses to standardized instruments and interviews were analyzed using machine learning algorithms. Subjects were re-interviewed approximately 30 days later, and their language was compared to the original language to determine the presence of suicidal ideation., Results: The results show that language characteristics used to classify suicidality at the initial encounter are still present in the speech 30 days later (AUC = 89% (95% CI: 85-95%), p < .0001) and that algorithms trained on the second interviews could also identify the subjects that produced the first interviews (AUC = 85% (95% CI: 81-90%), p < .0001)., Conclusions: This approach explores the stability of suicidal language. When using advanced computational methods, the results show that a patient's language is similar 30 days after first captured, while responses to standard measures change. This can be useful when developing methods that identify the data-based phenotype of a subject., (© 2020 The Authors. Suicide and Life-Threatening Behavior published by Wiley Periodicals LLC on behalf of American Association of Suicidology.)

Objective: People with epilepsy are at increased risk for mental health comorbidities. Machine-learning methods based on spoken language can detect suicidality in adults. This study's purpose was to use spoken words to create machine-learning classifiers that identify current or lifetime history of comorbid psychiatric conditions in teenagers and young adults with epilepsy., Materials and Methods: Eligible participants were >12 years old with epilepsy. All participants were interviewed using the Mini International Neuropsychiatric Interview (MINI) or the MINI Kid Tracking and asked five open-ended conversational questions. N-grams and Linguistic Inquiry and Word Count (LIWC) word categories were used to construct machine learning classification models from language harvested from interviews. Data were analyzed for four individual MINI identified disorders and for three mutually exclusive groups: participants with no psychiatric disorders, participants with non-suicidal psychiatric disorders, and participants with any degree of suicidality. Performance was measured using areas under the receiver operating characteristic curve (AROCs)., Results: Classifiers were constructed from 227 interviews with 122 participants (7.5 ± 3.1 minutes and 454 ± 299 words). AROCs for models differentiating the non-overlapping groups and individual disorders ranged 57%-78% (many with P < .02)., Discussion and Conclusion: Machine-learning classifiers of spoken language can reliably identify current or lifetime history of suicidality and depression in people with epilepsy. Data suggest identification of anxiety and bipolar disorders may be achieved with larger data sets. Machine-learning analysis of spoken language can be promising as a useful screening alternative when traditional approaches are unwieldy (eg, telephone calls, primary care offices, school health clinics)., (© 2019 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd.)

Objective: Delay to resective epilepsy surgery results in avoidable disease burden and increased risk of mortality. The objective was to prospectively validate a natural language processing (NLP) application that uses provider notes to assign epilepsy surgery candidacy scores., Methods: The application was trained on notes from (1) patients with a diagnosis of epilepsy and a history of resective epilepsy surgery and (2) patients who were seizure-free without surgery. The testing set included all patients with unknown surgical candidacy status and an upcoming neurology visit. Training and testing sets were updated weekly for 1 year. One- to three-word phrases contained in patients' notes were used as features. Patients prospectively identified by the application as candidates for surgery were manually reviewed by two epileptologists. Performance metrics were defined by comparing NLP-derived surgical candidacy scores with surgical candidacy status from expert chart review., Results: The training set was updated weekly and included notes from a mean of 519 ± 67 patients. The area under the receiver operating characteristic curve (AUC) from 10-fold cross-validation was 0.90 ± 0.04 (range = 0.83-0.96) and improved by 0.002 per week (P < .001) as new patients were added to the training set. Of the 6395 patients who visited the neurology clinic, 4211 (67%) were evaluated by the model. The prospective AUC on this test set was 0.79 (95% confidence interval [CI] = 0.62-0.96). Using the optimal surgical candidacy score threshold, sensitivity was 0.80 (95% CI = 0.29-0.99), specificity was 0.77 (95% CI = 0.64-0.88), positive predictive value was 0.25 (95% CI = 0.07-0.52), and negative predictive value was 0.98 (95% CI = 0.87-1.00). The number needed to screen was 5.6., Significance: An electronic health record-integrated NLP application can accurately assign surgical candidacy scores to patients in a clinical setting., (Wiley Periodicals, Inc. © 2019 International League Against Epilepsy.)

Racial disparities in the utilization of epilepsy surgery are well documented, but it is unknown whether a natural language processing (NLP) algorithm trained on physician notes would produce biased recommendations for epilepsy presurgical evaluations. To assess this, an NLP algorithm was trained to identify potential surgical candidates using 1097 notes from 175 epilepsy patients with a history of resective epilepsy surgery and 268 patients who achieved seizure freedom without surgery (total N = 443 patients). The model was tested on 8340 notes from 3776 patients with epilepsy whose surgical candidacy status was unknown (2029 male, 1747 female, median age = 9 years; age range = 0-60 years). Multiple linear regression using demographic variables as covariates was used to test for correlations between patient race and surgical candidacy scores. After accounting for other demographic and socioeconomic variables, patient race, gender, and primary language did not influence surgical candidacy scores (P > .35 for all). Higher scores were given to patients >18 years old who traveled farther to receive care, and those who had a higher family income and public insurance (P < .001, .001, .001, and .01, respectively). Demographic effects on surgical candidacy scores appeared to reflect patterns in patient referrals., (Wiley Periodicals, Inc. © 2019 International League Against Epilepsy.)

Background: Probabilistic assessments of clinical care are essential for quality care. Yet, machine learning, which supports this care process has been limited to categorical results. To maximize its usefulness, it is important to find novel approaches that calibrate the ML output with a likelihood scale. Current state-of-the-art calibration methods are generally accurate and applicable to many ML models, but improved granularity and accuracy of such methods would increase the information available for clinical decision making. This novel non-parametric Bayesian approach is demonstrated on a variety of data sets, including simulated classifier outputs, biomedical data sets from the University of California, Irvine (UCI) Machine Learning Repository, and a clinical data set built to determine suicide risk from the language of emergency department patients., Results: The method is first demonstrated on support-vector machine (SVM) models, which generally produce well-behaved, well understood scores. The method produces calibrations that are comparable to the state-of-the-art Bayesian Binning in Quantiles (BBQ) method when the SVM models are able to effectively separate cases and controls. However, as the SVM models' ability to discriminate classes decreases, our approach yields more granular and dynamic calibrated probabilities comparing to the BBQ method. Improvements in granularity and range are even more dramatic when the discrimination between the classes is artificially degraded by replacing the SVM model with an ad hoc k-means classifier., Conclusions: The method allows both clinicians and patients to have a more nuanced view of the output of an ML model, allowing better decision making. The method is demonstrated on simulated data, various biomedical data sets and a clinical data set, to which diverse ML methods are applied. Trivially extending the method to (non-ML) clinical scores is also discussed.

The lifetime of the τ lepton is measured using the process e+ e- → τ+ τ- , where both τ leptons decay to 3πν(τ). The result for the mean lifetime, based on 711  fb(-1) of data collected with the Belle detector at the ϒ(4S) resonance and 60  MeV below, is τ=(290.17±0.53(stat)±0.33(syst))×10(-15)  s. The first measurement of the lifetime difference between τ+ and τ- is performed. The upper limit on the relative lifetime difference between positive and negative τ leptons is |Δτ|/τ<7.0×10(-3) at 90% C.L.

We report measurements of B→χ(c1)γK and χ(c2)γK decays using 772×10(6) BB[over ¯] events collected at the Υ(4S) resonance with the Belle detector at the KEKB asymmetric-energy e(+)e(-) collider. Evidence of a new resonance in the χ(c1)γ final state is found with a statistical significance of 3.8σ. This state has a mass of 3823.1±1.8(stat)±0.7(syst) MeV/c(2), a value that is consistent with theoretical expectations for the previously unseen 1(3)D(2) cc[over ¯] meson. We find no other narrow resonance and set upper limits on the branching fractions of the X(3872)→χ(c1)γ and χ(c2)γ decays.

The cross section for ee+ e- → π+ π- J/ψ between 3.8 and 5.5 GeV is measured with a 967  fb(-1) data sample collected by the Belle detector at or near the Υ(nS) (n = 1,2,…,5) resonances. The Y(4260) state is observed, and its resonance parameters are determined. In addition, an excess of π+ π- J/ψ production around 4 GeV is observed. This feature can be described by a Breit-Wigner parametrization with properties that are consistent with the Y(4008) state that was previously reported by Belle. In a study of Y(4260) → π+ π- J/ψ decays, a structure is observed in the M(π(±)J/ψ) mass spectrum with 5.2σ significance, with mass M = (3894.5 ± 6.6 ± 4.5)  MeV/c2 and width Γ = (63 ± 24 ± 26)  MeV/c2, where the errors are statistical and systematic, respectively. This structure can be interpreted as a new charged charmoniumlike state.

We measure the branching fraction of B- → τ- ν(τ) using the full Υ(4S) data sample containing 772×10(6) BB pairs collected with the Belle detector at the KEKB asymmetric-energy e+ e- collider. Events with BB pairs are tagged by reconstructing one of the B mesons decaying into hadronic final states, and B- → τ- ν(τ) candidates are detected in the recoil. We find evidence for B- → τ- ν(τ) with a significance of 3.0 standard deviations including systematic errors and measure a branching fraction B(B- → τ- ν(τ))=[0.72(-0.25)(+0.27)(stat)±0.11(syst)]×10(-4).

We report the first evidence for the η(b)(2S) using the h(b)(2P)→η(b)(2S)γ transition and the first observation of the h(b)(1P)→η(b)(1S)γ and h(b)(2P)→η(b)(1S)γ transitions. The mass and width of the η(b)(1S) and η(b)(2S) are measured to be m(η(b)(1S))=(9402.4±1.5±1.8) MeV/c(2), m(η(b)(2S))=(9999.0±3.5(-1.9)(+2.8)) MeV/c(2), and Γ(η(b)(1S))=(10.8(-3.7-2.0)(+4.0+4.5)) MeV. We also update the h(b)(1P) and h(b)(2P) mass measurements. We use a 133.4 fb(-1) data sample collected at energies near the Υ(5S) resonance with the Belle detector at the KEKB asymmetric-energy e(+)e(-) collider.

We report the observation of two narrow structures in the mass spectra of the π(±)Υ(nS) (n=1, 2, 3) and π(±)h(b)(mP) (m=1, 2) pairs that are produced in association with a single charged pion in Υ(5S) decays. The measured masses and widths of the two structures averaged over the five final states are M(1)=(10,607.2±2.0)  MeV/c2, Γ(1)=(18.4±2.4)  MeV, and M(2)=(10,652.2±1.5)  MeV/c2, Γ(2)=(11.5±2.2)  MeV. The results are obtained with a 121.4  fb(-1) data sample collected with the Belle detector in the vicinity of the Υ(5S) resonance at the KEKB asymmetric-energy e+ e- collider.

Since brain damage in closed head injury of acceleration-deceleration type depends on the site of impact and on the course of the traumatising force, a reconstruction of these was attempted using CT. The study included 45 standard CT studies of adults with closed acceleration-deceleration head injuries. We selected 32 patients exclusively on the basis of CT findings of soft tissue contusion, skull fractures and cerebral contusions (coup and/or contrecoup, as well as other parenchymal lesions), while in the remaining 13 patients clinical findings were also used. On CT, the axial section of the skull was divided into 12 sections imitating the clock-face, to permit computerised graphic presentation of the direction of the traumatising force. Analysis of cerebral contusions on different CT studies in each case allowed location of "dominant coup" and "dominant contrecoup" lesions to be determined. The site of impact and the course of the traumatising force were reconstructed and graphically presented on the basis of these findings, supplemented with data on soft tissue contusions (present in 71% of cases), skull fractures (in 36%) and sometimes on other brain lesions. Comparison of the computerised graphic presentation of the site of impact and direction of the traumatic force and the location of lesions revealed a high correlation between them. In 80% of cases, the site of impact could be visualised only by CT. The acceleration force acted along the longer axis of the head (centroaxial or semioblique) in 87% and along the shorter axis in 13% of cases. Multiple lesions were found in 87% of cases. The lesions were most frequent in the frontal (51%) and temporal (26%) lobes. CT was very useful for reconstruction of the site of impact and of the course of the traumatising force in acceleration head injury. Data obtained by this procedure may have far-reaching prognostic and forensic implications.

Introduction: The delivery of radiofrequency (RF) energy through conventional catheter electrodes is often associated with coagulation necrosis at the tissue-electrode interface, with resultant impedance rise and limited lesion size. This study was performed to examine the effects of catheter tip cooling during RF delivery, to test the hypothesis that such cooling would decrease the likelihood if impedance rise and allow the creation of larger endomyocardial lesions., Methods and Results: The experiments were performed in eight open chest, anesthetized sheep. RF lesions were created within both ventricular chambers of each animal through a catheter tip that could be cooled with a saline perfusate. Assignment of cooled versus noncooled RF delivery to either ventricle was alternated from one animal to the next. In each set of experiments, lesion volumes relative to the mode of RF delivery were compared. The mean power delivered via the cooled electrode (22.04 +/- 4.51 W) was significantly higher than that delivered via the noncooled electrode (6.10 +/- 2.47 W; P < 0.001). The mean duration of RF delivery was 42.7 +/- 11.2 sec for noncooled lesions versus 49.2 +/- 6.8 sec for cooled lesions (P < 0.01). Mean lesion volume was 436.07 +/- 177.00 mm3 for noncooled RF delivery versus 1247.78 +/- 520.51 mm3 for cooled RF delivery (P < 0.001). This significantly larger lesion size with cooled RF delivery was associated with no instance of impedance rise in 27 attempts versus 11 impedance rises in 28 attempts with noncooled RF (P < 0.001)., Conclusions: Delivery of RF energy through a cooled catheter tip allows the creation of larger endomyocardial lesions by limiting the occurrence of impedance rise despite the delivery of greater energy. These observations suggest that, under certain conditions, resistive tissue heating at a distance from the site of current delivery may play an important role in RF ablation therapy.

Between August 1991 and December 1992, CT was performed on 154 patients who had suffered missile head injury during the war in the Republic of Croatia. In 54% CT was performed 1-24 h after injury, and in 27% follow-up CT was also obtained. The wounds were penetrating, tangential or perforating (45%, 34% and 21%, respectively). Haemorrhage was the most frequent lesion in the brain (84%). Follow-up CT evolution of haemorrhage, oedema, cerebritis, abscess, secondary vascular lesions, necrosis, encephalomalacia and hydrocephalus. The most dynamic changes occurred 7-14 days after injury. In 14% of cases, deep cerebral lesions were found in the corpus callosum, septum pellucidum periventricular region and pons, although bone and shell fragments were in a different part of the brain parenchyma. Such lesions were found in penetrating injuries only. CT proved very useful for assessing the extent and type of lesions. Although different mechanisms of brain damage in missile head injury are known, here they are, to the best of our knowledge, shown for the first time by CT.

A method which enables analysis of DNA from archival paraffin embedded normal and malignant brain tissue is described. The demonstration of a 317-bp long beta-actin DNA sequence by the polymerase chain reaction (PCR) was used to identify which fixation procedure, deparaffinization time and DNA extraction procedure would give the best results. Tissue specimens 1-39 years old were included in the experiments. Specimens fixed in either 10% formalin, Carnoy's or AMeX fixative were found to be best suited for subsequent analysis by PCR. Paraformaldehyde and acetone compromised amplification efficiency, while Bouin's fixed tissue gave uniformly negative results. Regardless of fixative used, PCR reaction had to be run through at least 40 cycles. Prolonged deparaffinization time and phenol/chloroform extraction of DNA did not influence DNA quality as a template for PCR reaction. Formalin fixed brain tumours can be successfully used for DNA/PCR analysis even if they are up to 39 years old.

A 14-year old boy was admitted for signs of heart failure and scapuloperoneal muscle weakness. He fulfilled the clinical, functional and diagnostic criteria for dilated cardiomyopathy. There was also a moderate increase in pulmonary vascular resistance. The immunohistochemical examination of the heart muscle revealed a slightly positive phytohemagglutinin reaction and minimal IgM deposits without complement. The electron microscopy examination disclosed increased numbers of abnormal mitochondria disrupting the usual cell structure; the mitochondria were of various sizes with irregular and abnormal structure of the cristae. The scapuloperoneal spinal muscular atrophy was mild and diagnosed according to clinical and electromyographic findings. Light microscope examination of the skeletal muscle revealed hypotrophic fibers. This patient is presumed to have postinflammatory mitochondriopathy and is currently being managed on low-dose digitalis, diuretics and captopril.

Computed tomographic analysis of lesions of the corpus callosum in 13 patients with "inner cerebral trauma" showed significant congruence of linear translation of acceleration and the topographic distribution of such lesions. This congruence permits computed tomography to be used to reconstruct the course of linear translation and the site of the main blow, which can be important for forensic use. The findings of even a small lesion in the corpus callosum indicates the need for further investigation of other structures which are usually involved in inner cerebral trauma, such as the hippocampus and brain stem.

A group of 42 patients with closed head injuries of acceleration type is reported. All of them were treated at various departments of the General Hospital in Maribor. The aim of investigation was to find out the diagnostic possibilities of head and brain damages in closed head injuries of acceleration type by the method of computed tomography (CT) and making comparison with the clinical picture of each individual patient. The patients were not investigated at random but were grouped in accordance with the direction of traumatic force action. The CT analyses were performed according to characteristic patterns of traumatic brain disease. The circumstances of head injury were reliable guidelines for devoting more attention to some brain areas. The density of each lesion was examined, too. The investigation has shown that lesions of "inner cerebral trauma" pattern (ICT-Grcević 1965) are visible on CT at least as often as coup-contrecoup lesions or even more. Only in cases with latero-lateral direction of traumatizing force (control group) coup-contrecoup lesions were found more frequently. Deep cerebral lesions of the latero-lateral type of acceleration head trauma were never or seldom encountered.

On computed tomography (CT) layers through the tumor the quality of picture depends upon the tumoral tissue structure, i.e., upon its histological appearance. In the formation of this histological appearance to a greater or lesser extent a series of various elements are involved, which are responsible for the density on CT scan, e.g., blood vessels, fatty tissue, calcium, connective tissue, collagen, etc. For some tumors the most important characteristic is specific organization of tissues in the volume, which can be clearly seen on high-quality CT scan. Possible limitations are due to technical capabilities of a CT apparatus. In our study we have obtained the most valuable results by analyzing the quality of picture of neurinomas on the so-called native CT scans, and by evaluating the way in which the picture of the tumoral tissue changed by opacification after an intravenous injection of contrast medium. Our study included 23 intracranial neurinomas. Twenty of them were neurinomas of the statoacoustic nerve, 1 of the orbit, 1 of the Gasser's gaglion and 1 was in the parasagittal region. For each particular tumor on histological specimens Antoni A and Antoni B types of tissue were semiquantitatively identified. In our study we have chosen only those cases for which we had a plenty of tumoral tissue suitable for semiserial analysis. The results of the histological analysis were compared with the results of the CT assessment of the tumoral tissue. Larger necrotic areas and cystic formations within tumoral tissue, which we could not analyze on our histological specimens, were excluded from the evaluation of CT scans of these tumors. Our results indicate a positive correlation between the histological appearance of neurinomas and their CT scans. This correlation between the histological properties of the tumoral tissue and the CT scan was almost 100% in those cases in which one histological type of neurinomas significantly prevailed. The prevalence of one type of neurinoma tissue was found in 74% of cases. In almost 61% of neurinomas in the histological appearance Antoni A type of tissue prevailed, what was clearly seen on CT in the form of hyperdense areas of the tumor. Antoni B type of tissue was histologically found in about 13% of neurinomas. These tumors had a marked hypodense picture of the tumoral tissue. The rest of neurinomas had mixed types of tissue. In this group of tumors we could not use the CT findings as an indicator in preoperative analysis of histopathological tissue characteristics.

Subdural hematoma (SDH) develops as a result of bleeding in the subdural space. According to nowadays accepted division, three groups of subdural hematomas can be differentiated: acute, subacute and chronic. The time elapsed from the moment of the occurrence of the hematoma to the moment when it was diagnosed is the main factor for determining the stage of SDH. However, for the above-mentioned types of SDH, this time differs depending on the author reporting it. Subdural hematoma is most often diagnosed by means of computerized tomography (CT). This method is safe and reliable, giving the exact diagnosis in more than 90% of cases. According to the basic principle of the concept of "living pathology", the knowledge of histological appearance of an investigated lesion is essential for the diagnostic interpretation of this lesion in neuroimaging methods. Very few authors studied the histological picture of subdural hematoma. The only structure which was histologically examined in details was the subdural neomembrane. Studies correlating histological picture of SDH and its appearance on CT scans have not been carried out until now. In this work such a correlation was made, and some regularities connecting these two methods were pointed out. Hyperdense picture of SDH on CT scans represents a hematoma containing almost only erythrocytes and erythrocyte-fibrin component being formed. Hypodense picture of SDH on CT scans represents a hematoma containing fibrin and inflammatory cells. Hematomas of mixed density on CT scan in all cases contained a neomembrane. Obvious histological differences between the mentioned types of subdural hematoma have led to the conclusion that chronic subdural hematoma is not the last stage of an "old" acute SDH. Chronic and acute subdural hematomas are different entities, considering their etiopathogenetic and clinical picture, and especially their CT and histopathological appearances.

A case of endotheliomatous type of meningioma arising from the falx and dura of the left anterior cranial fossa is reported. Numerous Rosenthal fibres were observed in the gliotic tissue around the meningioma. Authors present a brief data about the nature and occurrence of Rosenthal fibres. The occurrence of the pilocytic astrocytes and Rosenthal fibres in the gliotic tissue around slow-growing falx meningioma is an additional evidence for a peculiar type of changes in astrocytes of the long-standing gliosis of piloid type.

This study deals with the topographic distribution of traumatic lesions in the corpus callosum within the pattern of "inner cerebral trauma". Material for neuropathological investigations consisted of 49 brains of patients who suffered closed head injury of the acceleration type with linear translation of acceleration acting along the longer axis of the head. As a control, 5 brains were investigated in which the direction of linear translation was latero-lateral. The patients were divided into 4 groups according to reconstruction of the traumatic event and ascertaining of the direction of the linear translation of acceleration: (1) Medial blow (antero-posterior and postero-anterior), (2) Semioblique left blow (frontal left-occipital right and occipital right-frontal left), (3) Semioblique right blow (frontal right-occipital left and occipital left-frontal right), (4) Vertex-base direction. Fornix and septum pellucidum, as anatomically adjacent structures were also investigated. The results showed that the lesions in the corpus callosum were related to the pattern of "inner cerebral trauma" (ICT) (Grcević 1982), and the biomechanical conditions originally proposed by Lindenberg. These investigations, carried out on a representative material of 54 brains and by subserial histological analysis of the entire corpus callosum, produced a new evidence for our previous preliminary observations about the correlation of the shape and distribution of the periaxial lesions of the pattern of "inner cerebral trauma" and direction of the linear translation of accelerating forces. Our present investigations showed a striking congruence between the pattern of lesions within the corpus callosum in such type of cerebral trauma and the course of linear translation of acceleration. Furthermore, this study proved the importance of topographic interpretation of lesions in the corpus callosum in vivo i.e. by the CT and MRI examinations, because they may serve as a significant information for reconstruction of the biomechanical conditions of the injury which can be of importance for early therapeutical strategy and forensic interpretations of the accident. By using silver impregnation-techniques for histopathological study of the corpus callosum, we could confirm our previously expressed concept on "focalized" and not "diffuse" axonal injures in ICT (Grcević 1988). In this study corpus callosum served as a representative anatomical structure for axonal studies showing that axonal lesions strictly follow the pattern of tissue lesions with definite focal principle.

Autosomal dominant motor and sensory neuropathy with liability to pressure palsies was studied in three members of the same family. Only one of two monozygotic twin sisters was clinically affected. She developed unilateral peroneal palsy twenty minutes following local pressure. Electromyography revealed a weak intermediate innervation pattern with very rapid action potentials in the right anterior lower leg muscle. A 25-70 per cent reduction of motor and sensory conduction velocity was recorded in the clinically unaffected twin sister and in the father. The electrophysiological findings in the mother were normal. The sural nerve biopsy revealed "sausage-like" formations. The palsy persisted for two months and disappeared after eight weeks of fluocortolon treatment. It is possible that the myelin sheaths acted as antigen.

Estrogen (ER), progestin (PGR), and androgen (AR) receptors were assayed in cytosols prepared from 38 various intracranial tumors. The receptors were in the following proportions (number of receptor-positive/number of tumors examined): meningiomas were positive for PGR (4/6) and AR (2/5); glioblastomas were also positive for PGR (3/21) and AR (7/21); astrocytomas were positive only for PGR (4/5); and oligodendrogliomas only for AR. In two hamartomas AR was present, while in one chordoma both PGR and AR were present. In this latter tumor ER were not assayed due to insufficient material. The receptors were present in concentrations between 10 and 20 fmol/mg protein. Exceptions were two meningiomas and a chordoma with a high concentration of PGR and AR. Our results support the notion that a proportion of intracranial tumors contains sex steroid receptors, and some of these tumors might be hormonally dependent.

In this paper three cases with peculiar disease of the cerebellar cortex are presented. The disease is well known as L'hermitte-Duclos disease (LDD), but nowadays it is also called "dysplastic granulo-molecular hypertrophy of the cerebellar cortex" according to the modern theories of its etiology. Curious changes of the cerebellum in LDD are probably the result of a combination of delayed migration of Purkinje cells which occurs during the intrauterine period of cerebellar development, and of reactive hypertrophy of granular cells. Presented cases of L'hermitte-Duclos disease are, according to the literature, more or less typical by their clinical appearance and by their histological findings. Our electromicroscopical data support the findings of those authors who found only axodendritic sinapses on the hypertrophied ganglion cells, according to the idea that those cells are originally granular cells of the cerebellar cortex. CT scan of all our patients revealed parallel, worm-like, hyperdense formations in a hypodense process of the cerebellum. The same picture existed in some other described patients, but attention was paid to it. We consider this picture on the computed tomography to be characteristic, if not even patognomonic for the L'hermitte-Duclos disease, which gives us an entirely new scientific contribution in the process of the diagnosis of this peculiar disease.

A case of medullomyoblastoma, arising in the cerebellar vermis of a five-year-old boy, is presented. The light and electron microscopic features of the tumor are described and compared to other published cases. Within typical medulloblastoma areas light microscopy revealed signs of neuroblastic differentiation thus indicating its neuroepithelial origin. No signs of glial differentiation were found. Myogenic tumor component formed irregular, distinctly separated islands although some intermingling of two cell types was observed at the borders between the two zones. Within the myogenic tumor component, electron microscopy revealed rhabdomyoblastic elements in different stages of differentiation, frequently forming small groups encompassed by the same basement membrane. No indication of a common host cell for two cell lines was observed. No mesenchymal elements other than muscle, and no indication of possible teratoid origin of this tumor were observed. Considering histogenesis of muscle elements within this neuroepithelial tumor, the origin from multipotential neural-crest-derived ectomesenchymal cells seems the most appropriate one.

A group of 110 patients (68 male and 42 female) with cerebral metastases, treated at the Institute of Clinical Oncology and Radiotherapy, Zagreb University School of Medicine, during the period 1978-1984, were included in the study. Most patients were aged 50-60 years. Out of 110 patients, 52 were treated by radiotherapy and 58 by radiotherapy plus chemotherapy. Metastases from the bronchus carcinoma, breast carcinoma, melanoma and gastrointestinal carcinoma were present in 59%, 21.8% and 4.6% of patients, respectively. In 1.8% metastases from hypernephroma and in 3.6% from other malignant tumors were observed. In 4.6% cases, the origin of metastases could not be identified. Fifty-two out of 110 patients were treated by radiotherapy alone. They received 3000 cGy in 8-10 fractions, to the whole brain, with two parallel opposed fields. Fifty-eight out of 110 patients treated with radiotherapy and chemotherapy were given the same radiotherapeutic treatment. Chemotherapeutically, they were treated with BCNU and CCNU with or without Vincristin in standard doses. In the group of 52 patients treated by radiotherapy alone the median survival was six months (1-16 months), i.e. the same as in the group treated by both radiotherapy and chemotherapy (1-26 months).

A case of neuro-cutaneous melanoma, in the course of which a bifocal melanoma of the cerebral hemisphere had developed, was used as a natural model for the study of the relation between tumorous and non-tumorous elements. The need is pointed out for the definition of such a cutaneous-meningeal syndrome before the development of a neoplasm. As tumours develop from the cells defining leptomeningeal melanosis, the possibility of a neuroradiological diagnosis of this process is accentuated, primarily by a minute examination of the sites characteristic of the disease. A premorbid detection of all such cases is imperative in order to introduce an early anti-tumour treatment.

The authors present a case of herpes-simplex encephalitis in a 58-year-old woman. The disease had a biphasic course and lasted 87 days. The clinical picture, laboratory data and the pathologic-anatomical changes of the disease are discussed. The authors also emphasize the importance of the encephalitic process in the brain stem, which can make the clinical picture even more complicated because of the appearance of respiratory disturbances.

Since the first report of a defibrillation attempt with an intracardiac catheter electrode nearly 30 years ago, investigators have developed implantable electrode systems consisting of metal disks, endocardial catheters, and epicardial patches. These early efforts demonstrated the feasibility of low-energy reversion of ventricular tachyarrhythmias, and also provided some insight into the mechanisms of fibrillation and defibrillation. This review describes the evolution of implantable defibrillator electrode systems. Early investigators attempted defibrillation with submuscularly implanted metal disks or a disk electrode paired with an endocardial catheter electrode. Electrode design emphasis turned to transvenous catheter systems with electrodes placed in the right ventricle and right atrium. A more successful configuration placed the proximal electrode in the superior vena cava. In an effort to ensure proper placement of the distal electrode in humans, the catheter was replaced with an epicardial patch. More recently, a combination of electrodes and multiple pulses has substantially reduced the energy required to defibrillate. Effective electrode systems that can convert lethal arrhythmias with a minimum of energy will aid in making implantable cardioverters and defibrillators the therapy of choice in patients at high risk of sudden coronary death.

The authors present results of histological and electron-microscopical investigations of muscular tissue obtained by biopsy from a 20-year-old man, who showed typical features of Marfan's syndrome associated with slowly progressive muscular weakness. The muscle showed a peculiar combination of centronuclear myopathy with hypotrophy of type I fibres and the so called "fingerprint myopathy". The combination of such myopathic phenomena with Marfan's syndrome presents a unique and hitherto undescribed condition.

The authors report on 79 cases of primary brain sarcomas, correlating clinical and pathoanatomical data. In the material analyzed the group of gliosarcomas prevailed, while the smallest group were rhabdomyosarcomas. In some groups the topic predilection did not show any specific characteristics, except that there were more gliosarcomas occurring in the parietal lobe. The symptoms lasted 1-3 months before operation.