Terms Of Positive Predictive Value Research Articles

Benefits of Premaquick® Combined Detection of IL-6/Total IGFBP-1/Native IGFBP-1 to Predict Preterm Delivery.

We conducted a prospective double-blind study to compare two vaginal diagnostic methods in singleton pregnancies with threatened preterm labor (TPL) at the University Hospital of Clermont-Ferrand (France) from August 2018 to December 2020. Our main objective was to compare the diagnostic capacity at admission, in terms of positive predictive value (PPV) and negative predictive value (NPV), of Premaquick® (combined detection of IL-6/total IGFBP-1/native IGFBP-1) and QuikCheck fFN™ (fetal fibronectin) for delivery within 7 days in cases of TPL. We included 193 patients. Premaquick® had a sensitivity close to 89%, equivalent to QuikCheck fFN™, but a higher statistical specificity of 49.5% against 38.6% for QuikCheck fFN™. We found no superiority of Premaquick® over QuickCheck fFN™ in terms of PPV (6.6% vs. 7.9%), with NPV being equivalent in predicting childbirth within 7 days in cases of TPL (98.6% vs. 98.9%). Nevertheless, the combination of positive native and total IGFBP-1 and the combination of all three positive markers were associated with a higher PPV. Our results, though non-significant, support this combined multiple-biomarker approach to improve testing in terms of predictive values.

Development of machine learning models for the surveillance of colon surgical site infections

Conventional surgical site infection (SSI) surveillance is labour-intensive. We aimed to develop machine learning (ML) models for the surveillance of SSIs for colon surgery and to assess whether the ML could improve surveillance process efficiency. This study included cases who underwent colon surgery at a tertiary center between 2013 and 2014. Logistic regression and four ML algorithms including random forest (RF), gradient boosting (GB), and neural networks (NNs) with or without recursive feature elimination (RFE) were first trained on the entire cohort, and then re-trained on cases selected based on a previous rule-based algorithm. We assessed model performance based on the area under the curve (AUC), sensitivity, and positive predictive value (PPV). The estimated proportion of reduction in workload for chart review based on the ML models was evaluated and compared with the conventional method. At a sensitivity of 95%, the NN with RFE using 29 variables had the best performance with an AUC of 0.963 and PPV of 21.1%. When combining both the rule-based algorithm and ML algorithms, the NN with RFE using 19 variables had a higher PPV (28.9%) than with the ML algorithm alone, which could decrease the number of cases requiring chart review by 83.9% compared with the conventional method. We demonstrated that ML can improve the efficiency of SSI surveillance for colon surgery by decreasing the burden of chart review while providing high sensitivity. In particular, the hybrid approach of ML with a rule-based algorithm showed the best performance in terms of PPV.

Therapy Defining at Initial Diagnosis of Primary Brain Tumor-The Role of 18F-FET PET/CT and MRI.

Primary malignant brain tumors are heterogeneous and infrequent neoplasms. Their classification, therapeutic regimen and prognosis have undergone significant development requiring the innovation of an imaging diagnostic. The performance of enhanced magnetic resonance imaging depends on blood-brain barrier function. Several studies have demonstrated the advantages of static and dynamic amino acid PET/CT providing accurate metabolic status in the neurooncological setting. The aim of our single-center retrospective study was to test the primary diagnostic role of amino acid PET/CT compared to enhanced MRI. Emphasis was placed on cases prior to intervention, therefore, a certain natural bias was inevitable. In our analysis for newly found brain tumors 18F-FET PET/CT outperformed contrast MRI and PWI in terms of sensitivity and negative predictive value (100% vs. 52.9% and 36.36%; 100% vs. 38.46% and 41.67%), in terms of positive predictive value their performance was roughly the same (84.21 % vs. 90% and 100%), whereas regarding specificity contrast MRI and PWI were superior (40% vs. 83.33% and 100%). Based on these results the superiority of 18F-FET PET/CT seems to present incremental value during the initial diagnosis. In the case of non-enhancing tumors, it should always be suggested as a therapy-determining test.

Explainable Computational Intelligence Model for Antepartum Fetal Monitoring to Predict the Risk of IUGR

Intrauterine Growth Restriction (IUGR) is a restriction of the fetus that involves the abnormal growth rate of the fetus, and it has a huge impact on the new-born’s health. Machine learning (ML) algorithms can help in early prediction and discrimination of the abnormality of the fetus’ health to assist in reducing the risk during the antepartum period. Therefore, in this study, Random Forest (RF), Support Vector Machine (SVM), K Nearest Neighbor (KNN) and Gradient Boosting (GB) was utilized to discriminate whether a fetus was healthy or suffering from IUGR based on the fetal heart rate (FHR). The Recursive Feature Elimination (RFE) method was used to select the significant feature for the classification of fetus. Furthermore, the study Explainable Artificial Intelligence (EAI) was implemented using LIME and SHAP to generate the explanation and to add comprehensibility in the proposed models. The experimental results indicate that RF achieved the highest accuracy (0.97) and F1-score (0.98) with the reduced set of features. However, the SVM outperformed it in terms of Positive Predictive Value (PPV) and specificity (SP). The performance of the model was further validated using another dataset and found that it outperformed the baseline studies for both the datasets. The proposed model can aid doctors in monitoring fetal health and enhancing the prediction process.

Child Maltreatment in Lebanon: Prevalence, Screening and Upgrade of ESCAPE

Child maltreatment is a violation of human rights and includes all forms of abuse against children under 18 years of age. It is the second cause of death in children after accidental trauma and remains an active problem worldwide. In Lebanon, little work is being done nationwide to provide and explain the available screening and reporting tools for child maltreatment. In this study, we assess the diagnostic accuracy of the ESCAPE instrument and its modified version (including one additional question “<i>Did parents/caregivers take adequate measures?</i>”) in screening child maltreatment in emergency departments, and calculate the prevalence of child abuse in Lebanon. The present diagnostic accuracy study is performed over 16 months using a convenience sampling on children under the age of 18 who presented to Lebanese emergency departments. Confirmation of maltreatment by the child protection team was considered as the cornerstone to diagnosis. Screening performance characteristics of ESCAPE and ESCAPE-modified were calculated using GPower v3.1 and SPSS v22 soft wares. 411 children with the mean age of 5.7 years were screened (38% girls). 47 children were suspected victims of child abuse (11.4%). 13 children (3.2%) were confirmed victims of child maltreatment. Sensitivity, specificity, positive and negative predictive values of this test, with 95% confidence intervals, were 100% (82.7 – 100), 91.5% (88.4 – 93.9), 27.7% (16.5 – 41.5), and 100% (99.3 – 100) respectively. The additional question added to ESCAPE did not alter the diagnostic performance of the test but was superior in terms of positive predictive value without losing in terms of negative predictive value. In this study, ESCAPE was proven to be an appropriate screening instrument for identifying children at high risk of child abuse, with excellent accuracy. The additional question of “<i>Did parents/caregivers take adequate measures?</i>” represents a simpler tool with a higher positive predictive value, to be used after a positive ESCAPE screening to help confirm or reject the diagnosis of child maltreatment in case of doubt.

An anti-phishing model based on similarity measurement

Phishing has represented a more noteworthy danger to clients. In the current work, the authors attempted to build up a powerful anti-phishing technique based on hybrid similarity approach combining Cosine and Soft Cosine similarity that measures the resemblance between user query and database. The proposed similarity hybrid is also evaluated against another similarity hybrid comprising of Cosine and Jaccard similarity measure so as to validate the proposed work. Both hybrid similarities are separately fed to validation layer of feed forward back propagation neural network (FFBPNN) to predict phishing and legitimate websites. The performance of the proposed work is evaluated against data set comprising of 3,000 sample files in terms of positive predictive value (PPV), true positive rate (TPR) and F-measure. The comparative analysis demonstrated that the anti-phishing model using proposed similarity hybrid outperformed the cosine and Jaccard similarity hybrid with 0.233%, 0.2833% and 0.258% higher PPV, TPR and F-measure, respectively.

Can Ensemble Machine Learning Improve the Accuracy of Severe Maternal Morbidity Screening in a Perinatal Database?

Severe maternal morbidity (SMM) is an important maternal health indicator, but existing tools to identify SMM have substantial limitations. Our objective was to retrospectively identify true SMM status using ensemble machine learning in a hospital database and to compare machine learning algorithm performance with existing tools for SMM identification. We screened all deliveries occurring at Magee-Womens Hospital, Pittsburgh, PA (2010-2011 and 2013-2017) using the Centers for Disease Control and Prevention list of diagnoses and procedures for SMM, intensive care unit admission, and/or prolonged postpartum length of stay. We performed a detailed medical record review to confirm case status. We trained ensemble machine learning (SuperLearner) algorithms, which "stack" predictions from multiple algorithms to obtain optimal predictions, on 171 SMM cases and 506 non-cases from 2010 to 2011, then evaluated the performance of these algorithms on 160 SMM cases and 337 non-cases from 2013 to 2017. Some SuperLearner algorithms performed better than existing screening criteria in terms of positive predictive value (0.77 vs. 0.64, respectively) and balanced accuracy (0.99 vs. 0.86, respectively). However, they did not perform as well as the screening criteria in terms of true-positive detection rate (0.008 vs. 0.32, respectively) and performed similarly in terms of negative predictive value. The most important predictor variables were intensive care unit admission and prolonged postpartum length of stay. Ensemble machine learning did not globally improve the ascertainment of true SMM cases. Our results suggest that accurate identification of SMM likely will remain a challenge in the absence of a universal definition of SMM or national obstetric surveillance systems.

Automated Radiology Alert System for Pneumothorax Detection on Chest Radiographs Improves Efficiency and Diagnostic Performance.

We aimed to set up an Automated Radiology Alert System (ARAS) for the detection of pneumothorax in chest radiographs by a deep learning model, and to compare its efficiency and diagnostic performance with the existing Manual Radiology Alert System (MRAS) at the tertiary medical center. This study retrospectively collected 1235 chest radiographs with pneumothorax labeling from 2013 to 2019, and 337 chest radiographs with negative findings in 2019 were separated into training and validation datasets for the deep learning model of ARAS. The efficiency before and after using the model was compared in terms of alert time and report time. During parallel running of the two systems from September to October 2020, chest radiographs prospectively acquired in the emergency department with age more than 6 years served as the testing dataset for comparison of diagnostic performance. The efficiency was improved after using the model, with mean alert time improving from 8.45 min to 0.69 min and the mean report time from 2.81 days to 1.59 days. The comparison of the diagnostic performance of both systems using 3739 chest radiographs acquired during parallel running showed that the ARAS was better than the MRAS as assessed in terms of sensitivity (recall), area under receiver operating characteristic curve, and F1 score (0.837 vs. 0.256, 0.914 vs. 0.628, and 0.754 vs. 0.407, respectively), but worse in terms of positive predictive value (PPV) (precision) (0.686 vs. 1.000). This study had successfully designed a deep learning model for pneumothorax detection on chest radiographs and set up an ARAS with improved efficiency and overall diagnostic performance.

Epileptogenic zone detection in MRI negative epilepsy using adaptive thresholding of arterial spin labeling data

Drug-resistant epilepsy is a diagnostic and therapeutic challenge, mainly in patients with negative MRI findings. State-of-the-art imaging methods complement standard epilepsy protocols with new information and help epileptologists to increase the reliability of their decisions. In this study, we investigate whether arterial spin labeling (ASL) perfusion MRI can help localize the epileptogenic zone (EZ). To that end, we developed an image processing method to detect the EZ as an area with hypoperfusion relative to the contralateral unaffected side, using subject-specific thresholding of the asymmetry index in ASL images. We demonstrated three thresholding criteria (termed minimal product criterion, minimal distance criterion, and elbow criterion) on 29 patients with MRI-negative epilepsy (age 32.98 ± 10.4 years). The minimal product criterion showed optimal results in terms of positive predictive value (mean 0.12 in postoperative group and 0.22 in preoperative group) and true positive rate (mean 0.71 in postoperative group and 1.82 in preoperative group). Additionally, we found high accuracy in determining the EZ side (mean 0.86 in postoperative group and 0.73 in preoperative group out of 1.00). ASL can be easily incorporated into the standard presurgical MR protocol, and it provides an additional benefit in EZ localization.

Improvement of Neoantigen Identification Through Convolution Neural Network.

Accurate prediction of neoantigens and the subsequent elicited protective anti-tumor response are particularly important for the development of cancer vaccine and adoptive T-cell therapy. However, current algorithms for predicting neoantigens are limited by in vitro binding affinity data and algorithmic constraints, inevitably resulting in high false positives. In this study, we proposed a deep convolutional neural network named APPM (antigen presentation prediction model) to predict antigen presentation in the context of human leukocyte antigen (HLA) class I alleles. APPM is trained on large mass spectrometry (MS) HLA-peptides datasets and evaluated with an independent MS benchmark. Results show that APPM outperforms the methods recommended by the immune epitope database (IEDB) in terms of positive predictive value (PPV) (0.40 vs. 0.22), which will further increase after combining these two approaches (PPV = 0.51). We further applied our model to the prediction of neoantigens from consensus driver mutations and identified 16,000 putative neoantigens with hallmarks of ‘drivers’.

HeMA: A hierarchically enriched machine learning approach for managing false alarms in real time: A sepsis prediction case study

Early detection of sepsis can be life-saving. Machine learning models have shown great promise in early sepsis prediction when applied to patient physiological data in real-time. However, these existing models often under-perform in terms of positive predictive value, an important metric in clinical settings. This is especially the case when the models are applied to data with less than 50% sepsis prevalence, reflective of the incidence rate of sepsis on the floor or in the ICU. In this study, we develop HeMA, a hierarchically enriched machine learning approach for managing false alarms in real time, and conduct a case study for early sepsis prediction. Specifically, we develop a two-stage framework, where a first stage machine learning model is paired with statistical tests, particularly Kolmogorov-Smirnov tests, in the second stage, to predict whether a patient would develop sepsis. Compared with machine learning models alone, the framework results in an increase in specificity and positive predictive value, without compromising F1 score. In particular, the framework shows improved performance when applied to data with 50% and 25% sepsis prevalence, collected from a large hospital system in the US, resulting in up to 18% and 7% increase in specificity and positive predictive value, respectively. Despite the significant improvements observed, and although F1 score is not negatively affected, because of the up to 6% decrease in sensitivity, further improvements and pilot studies may be necessary before deploying the framework in a clinical setting. Finally, external validation conducted using a publicly available dataset produces similar results, validating that the proposed framework is generalizable.

The Trigger Tool Method for Routine Pharmacovigilance: A Retrospective Cohort Study of the Medical Records of Hospitalized Geriatric Patients.

The main objective was to assess the feasibility of the trigger tool method for the retrospective detection of adverse drug reactions (ADRs) in the Rennes University Hospital. The secondary objective was to describe the performance of the method in terms of positive predictive values (PPVs) and severity or preventability of ADRs. Using the Rennes University Hospital clinical data warehouse, pharmacovigilance experts performed a retrospective review of a random sample of 30 inpatient hospital medical records per month using the triggers "fall" and "delirium" to identify related ADRs among patients 65 years and older in 2018 in the geriatrics department. Using the Z test, we compared the proportion of medical records with a positive (identified) trigger related to an ADR, which were reviewed within 20 minutes using the reference of 50% reviewed within 20 minutes. Among the 355 medical records reviewed, 222 had at least 1 trigger and 98 at least 1 related ADR. Among the 222 positive trigger medical records, 99.6% were reviewed in under 20 minutes (P < 0.001). The pharmacovigilance assessment took 3 months. The PPVs reached 53.9% (46.0%-61.7%) for falls and 21.0% (14.3%-27.5%) for delirium. Among the ADRs, 80% were serious and 53% were preventable. Given the low PPV of the triggers used and the considerable need for technical and human resources, the trigger tool method cannot be used as a routine tool at the pharmacovigilance center. However, it could be implemented occasionally for specific purposes such as monitoring the impact of risk minimization measures to prevent ADRs.

High-resolution computed tomography as a non-invasive imaging biomarker for interstitial lung diseases: a tertiary center study

Background: Interstitial lung diseases (ILD) are a heterogeneous group of non-neoplastic disorders resulting from damage to the lung parenchyma by varying patterns of inflammation and fibrosis. With high-resolution computed tomography (HRCT) the pattern of lung damage can be mapped accurately which may help to identify specific ILD.Methods: 65 diagnosed cases of ILD by HRCT who were admitted to a tertiary care chest hospital, formed the study group. All these patients also underwent histopathological confirmation as per hospital protocol. The study was done over a period from August 2016 to July 2019. Clinical details, chest x-ray, HRCT and histopathological data was collected and analysed using 2x2 table for detecting sensitivity, specificity, positive predictive value (PPV) and negative predictive values (NPV).Results: For diagnosing ILD like acute interstitial pneumonia (AIP), LIP and RB ILD the HRCT fared equally well in diagnostic utility as compared to histopathological examination. But in certain conditions like non-specific interstitial pneumonia (NSIP) the HRCT performed poorly in terms of PPV as compared to gold standard histopathology. In Bronchiolitis obliterans organizing pneumonia (BOOP) and usual interstitial pneumonia (UIP) again the HRCT performed fairly well as compared to gold standard.Conclusions: HRCT shows good correlation with histopathological diagnosis in identifying a various subtype of ILD and may thus serve a useful non-invasive, imaging biomarker not only for diagnosing a particular ILD but for prognostication and response to treatment.

Comparison of laboratory threshold criteria in drug-induced liver injury detection algorithms for use in pharmacovigilance.

For the purpose of pharmacovigilance, we sought to determine the best performing laboratory threshold criteria to detect drug-induced liver injury (DILI) in the electronic medical records (EMR). We compared three commonly used liver chemistry criteria from the DILI expert working group (DEWG), DILI network (DILIN), and Council for International Organizations of Medical Sciences (CIOMS), based on hospital EMR for years 2010 and 2011 (42 176 admissions), using independent medical record review. The performance characteristics were compared in terms of sensitivity, specificity, positive predictive value (PPV), negative predictive value, accuracy, F-measure, and area under the receiver operating characteristic curve (AUROC). DEWG had the highest PPV (5.5%, 95% CI: 4.1%-7.2%), specificity (97.0%, 95% CI: 96.8%-97.2%), accuracy (96.8%, 95% CI: 96.6%-97.0%) and F-measure (0.099). CIOMS had the highest sensitivity (74.0%, 95% CI: 64.3%-82.3%) and AUROC (85.2%, 95% CI: 80.8%-89.7%). Besides the laboratory criteria, including additional keywords in the classification algorithm improved the PPV and F-measure to a maximum of 29.0% (95% CI: 22.3%-36.5%) and 0.379, respectively. More stringent criteria (DEWG and DILIN) performed better in terms of PPV, specificity, accuracy and F-measure. CIOMS performed better in terms of sensitivity. An algorithm with high sensitivity is useful in pharmacovigilance for detecting rare events and to avoid missing cases. Requiring at least two abnormal liver chemistries during hospitalization and text-word searching in the discharge summaries decreased false positives without loss in sensitivity.

Nucleus Segmentation of Cervical Cytology Images Based on Depth Information

In cervical cancer screening, accurate segmentation of cervical nucleus is a key part in the early diagnosis of cervical cancer. However, the cervical nucleus segmentation faces many great challenges owing to the overlapping cervical cells, uneven staining and poor contrast of cervical cytology smear images. In this paper, a tree domain structure and screening algorithm based on depth-first searching strategy are proposed to obtain candidate nucleus regions according to the annular clustering characteristics of nucleus depth information in cervical cytology images. Then, the candidate nucleus regions are finely segmented with an iterative level set algorithm based on adaptive radius morphological dilation. Experimental results are evaluated on the ISBI2015 public dataset. The performance of the proposed nucleus segmentation algorithm is higher than that of the state-of-the-art methods in terms of positive predictive value, negative predictive value, precision, recall of the cervical nucleus segmentation.

Evaluation of trigger tool method for adverse drug reaction reporting by nursing staff at a tertiary care teaching hospital

Background: To sensitize nurses about Trigger Tool Method (TTM) and to evaluate the impact of TTM on adverse drug event (ADE) reporting by nurses at a tertiary care teaching hospital in India.Methods: This was prospective, interventional, single center study conducted among nursing health professionals of Civil Hospital Ahmedabad (CHA) posted in Medicine Department. They were sensitized about ADE reporting, pharmacovigilance, methods of ADRs reporting and details about TTM. Also, a list of 17 triggers was prepared by the investigator and given to nurses. They were educated to report ADEs using TTM. At the initiation and end of study, questionnaires were given to evaluate knowledge, attitude and practice of ADR reporting among participant nurses. All triggers and ADEs reported were analyzed in terms of association between them, effectiveness of trigger in detecting an ADR and in terms of Positive Predictive Value (PPV). Reported ADRs were also assessed for causality, severity and preventability.Results: A total 758 patients were admitted during the study period in the respective medicine department. List of 17 triggers consists of 9 drug triggers (DT), 1 laboratory trigger (LT) and 7 patient triggers (PT). Of these 17 triggers, 14 triggers were identified by nurses. These 14 triggers were noticed 130 times. These included DT (100 times), LT (0 times) and PT (30 times). Of the various triggers observed, 7 DT and 4 PT were related to ADRs. Hence, 11 triggers (64.70%) were positive (related to ADRs), out of 17 total triggers under evaluation. 21 ADRs were observed using TTM by nurses.Conclusions: The TTM helps to detect and report ADRs by nurses. Educational interventions about TTM help in better detection and reporting of ADRs.

Parallel repulsive logic regression with biological adjacency.

Logic regression, an extension of generalized linear models with Boolean combinations of binary variables as predictors, is a useful tool in exploring interactions among single-nucleotide polymorphisms (SNPs) in genome-wide association studies. However, since the search space defined by all possible combinations of SNPs, their complements, and logical operators in Boolean expressions can be exceedingly large in such studies, objective function optimization is slow and likely to be trapped in many local solutions, resulting in model over-fitting. We introduce a new search algorithm, parallel repulsive logic regression (PRLR), to efficiently estimate parameters of a logic regression to find a best model within the large space of SNP interactions by incorporating: (i) relevant biological adjacency matrix between SNPs to define similarity of estimation paths or trees, which are derived from physical SNP positions on chromosomes and/or memberships in biological gene pathways; and (ii) two repulsive forces to counter the similarity between and within estimation paths considered in parallel, which are introduced as penalty terms in the objective function. We compare our method's performance for identifying biologically-meaningful SNP interactions through simulations and with real genetic-epidemiological data. PRLR's detection-accuracy measures outperform existing approaches, especially in terms of positive predictive value and sensitivity for detecting SNP-SNP interactions.

A Novel Tool for the Early Identification of Frailty in Elderly People: The Application in Primary Care Settings.

Frailty is a pre-disability condition in older persons providing a challenge to Health-Care Systems. Systematic reviews highlight the absence of a gold-standard for its identification. However, an approach based on initial screening by the General Practitioner (GP) seems particularly useful. On these premises, a 9-item Sunfrail Checklist (SC), was developed by a multidisciplinary group, in the context of European Sunfrail Project, and tested in the Community. - to measure the concordance between the judgments of frailty (criterion-validity): the one formulated by the GP, using the SC, and the one subsequently expressed by a Comprehensive Geriatric Assessment Team (CGA-Team); - to determine the construct-validity through the correspondence between some checklist items related to the 3 domains (physical, cognitive and social) and the three tools used by the CGA-Team; - to measure the instrument's performance in terms of positive predictive value (PPV) and negative predictive value (NPV). Cross-sectional study, with a final sample-size of 95 subjects. Two Community-Health Centers of Parma, Italy. Subjects aged 75 years old or more, with no disability and living in the community. We compared the screening capacity of the GP using the SC to that one of CGA-Team based on three tests: 4-meter Gait-Speed, Mini-Mental State Examination and Loneliness Scale. 95 subjects (51 women), with a mean age of 81±4 years were enrolled. According to GPs 34 subjects were frail; the CGA-Team expressed a frailty judgment on 26 subjects. The criterion-validity presented a Cohen's k of 0.353. Construct-validity was also low, with a maximum contingency-coefficient of 0.19. The analysis showed a PPV of 58.1% and a NPV equal to 84.6%. Our data showed a low agreement between the judgements of GP performed by SC and CGA-Team. However, the good NPV suggests the applicability of SC for screening activities in primary-care.

Predicting poorer health outcomes in older community-dwelling patients with multimorbidity: prospective cohort study assessing the accuracy of different multimorbidity definitions

PurposeMultimorbidity is commonly defined and measured using condition counts. The UK National Institute for Health Care Excellence Guidelines for Multimorbidity suggest that a medication-orientated approach could be used to identify...

Self-reported violent ideation and its link to interpersonal violence among offenders with mental disorders and general psychiatric patients

This study aims at comparing mentally disordered offenders and general psychiatric patients regarding violent ideation and at exploring its association with interpersonal violence. We recruited 200 detainees undergoing forensic psychiatric evaluation and 390 general psychiatric patients at discharge. At baseline, they were asked about violent ideation; at the 20-week follow-up, information about violent acts was gathered from crime conviction registry, interviews, and records. The lifetime prevalence of violent ideation was 32.5% for offenders and 35.6% for patients; the corresponding two-month prevalence was 22.5% and 21.0%, respectively. For the both samples combined, those with violent ideation in their lifetime were significantly more prone to commit violent acts during follow-up than those without such ideation, OR = 2.65. The same applied to the patient sample, OR = 3.41. In terms of positive predictive values, fewer than 25% of those with violent ideation committed violent acts. Contrary to our hypothesis, the prevalence of violent ideation did not differ significantly between offenders and patients. However, there was support for the hypothesized association between violent ideation and violent acts on a group level. On an individual level, the clinician should consider additional factors when assessing the risk for violent acts.