Deep neural network technique for automated detection of ADHD and CD using ECG signal.

BACKGROUND AND OBJECTIVE: Attention Deficit Hyperactivity problem (ADHD) is a common neurodevelopment problem in children and adolescents that can lead to long-term challenges in life outcomes if left untreated. Also, ADHD is frequently associated with Conduct Disorder (CD), and multiple research have found similarities in clinical signs and behavioral symptoms between both diseases, making differentiation between ADHD, ADHD comorbid with CD (ADHD+CD), and CD a subjective diagnosis. Therefore, the goal of this pilot study is to create the first explainable deep learning (DL) model for objective ECG-based ADHD/CD diagnosis as having an objective biomarker may improve diagnostic accuracy. METHODS: The dataset used in this study consist of ECG data collected from 45 ADHD, 62 ADHD+CD, and 16 CD patients at the Child Guidance Clinic in Singapore. The ECG data were segmented into 2 s epochs and directly used to train our 1-dimensional (1D) convolutional neural network (CNN) model. RESULTS: The proposed model yielded 96.04% classification accuracy, 96.26% precision, 95.99% sensitivity, and 96.11% F1-score. The Gradient-weighted class activation mapping (Grad-CAM) function was also used to highlight the important ECG characteristics at specific time points that most impact the classification score. CONCLUSION: In addition to achieving model performance results with our suggested DL method, Grad-CAM's implementation also offers vital temporal data that clinicians and other mental healthcare professionals can use to make wise medical judgments. We hope that by conducting this pilot study, we will be able to encourage larger-scale research with a larger biosignal dataset. Hence allowing biosignal-based computer-aided diagnostic (CAD) tools to be implemented in healthcare and ambulatory settings, as ECG can be easily obtained via wearable devices such as smartwatches.

Healthcare utilization and costs of singaporean youth with symptoms of depression and anxiety: results from a 2022 web panel.

BACKGROUND: There is evidence that the prevalence of depression and anxiety among youth is increasing and that these factors contribute to high healthcare costs and poor school performance. The goal of this study is to provide up-to-date estimates of the prevalence and economic burden of depression and anxiety among youth in Singapore. METHODS: Using an existing web panel, 991 parents filled out the PHQ-4 screener on behalf of 1,515 youth. 104 of these parents whose children (ages 4 to 21) had symptoms consistent with depression or anxiety filled out a full survey with questions on mental health symptoms, school absences, school performance, and healthcare utilization. The survey was fielded between April and June 2022. Publicly available prices were used to estimate the cost associated with the observed rates of health service use. FINDINGS: Based on parental responses, 11.7% (95% CI:10.2 - 13.5%) of youth had symptoms consistent with depression and 12.8% (95% CI:11.2 - 14.6%) had symptoms consistent with anxiety. In total, 16.2% (95% CI:14.5 - 18.3%) were reported to have symptoms consistent with at least one of these conditions. These youths missed an average of 190 (95% CI: 126-254) hours of school per year due to their mental health conditions and parents reported that school and daily activities performance was significantly degraded. Per capita annual healthcare costs averaged S$10,250 (95% CI: 7,150-13,350), with 64% of youth receiving emergency or inpatient services. In aggregate, annual costs associated with these conditions were estimated to be S$1.2 billion (95% CI:S$1.1bn - S$1.4bn). INTERPRETATION: Even with significant potential for underreporting, these results reveal concerning rates of Singaporean youth with symptoms consistent with depression or anxiety, many of whom remain untreated. Results also reveal the short-term economic burden caused by these symptoms and hint at longer-term consequences resulting from poor school performance. This study should represent a call to action for Singapore to address poor mental health among youth.

Prevalence and economic burden of depression and anxiety symptoms among Singaporean adults: results from a 2022 web panel.

BACKGROUND: Major depressive disorder (MDD) and generalized anxiety disorder (GAD) are leading causes of disability and premature mortality. At a global level, over 300 million people are estimated to suffer from major depressive disorders, equivalent to 4·4% of the world's population. Pandemic era stressors have increased rates for depression and anxiety by upwards of 25%. The goal of this study is to estimate the prevalence and economic burden of depression and anxiety symptoms in Singapore after the peak of the COVID-19 pandemic. METHODS: An existing web panel was queried between April 2022 and June 2022. Adult participants aged > 21 years old who screened positive for depression and anxiety symptoms based on the Patient Health Questionnaire-4 (PHQ-4) Screener were eligible for participation. Prevalence estimates were quantified by dividing the number of respondents who screened positive for these symptoms by the total number of respondents. Participants who screened positive were asked about healthcare utilization, days missed from work, and reduced productivity due to these symptoms. These values were then monetized and scaled based on prevalence and population counts to generate per capita and total annual costs. RESULTS: Two thousand three hundred forty-eight respondents filled out the PHQ-4 depression/anxiety screener on behalf of the 5,725 adults living in their households (including respondents themselves). Prevalence estimates were calculated based on the responses recorded for these 5,725 adults. 14.1% adults had symptoms consistent with depression and 15.2% had symptoms consistent with anxiety. In total, 20.0% may experience symptoms consistent with at least one of these two conditions, yet approximately half reported never being formally diagnosed. 350 respondents screened positive for depression or anxiety symptoms and thus were eligible to fill out the healthcare utilization, presenteeism, and absenteeism survey. Direct annual healthcare costs due to depression and anxiety symptoms averaged Singapore dollar (SGD) $1,050 for these respondents. The employed subset (n = 304) missed an extra 17.7 days of work on average per year, which translates to SGD $4,980 per worker. These workers also reported being ~ 40% less productive at work, which equates to SGD $28,720 in economic losses annually. In total, these symptoms caused SGD $15.7 billion in increased costs. Presenteeism accounts for 81.6% of this total (SGD $12.8 billion), absenteeism for 14.2% (SGD $2.3 billion) and healthcare accounts for 4.2% (SGD $0.7 billion). CONCLUSIONS: The health and economic burden associated with depression and anxiety symptoms is large in Singapore, representing 2.9% of Singapore's gross domestic product (GDP). Employers and governments should look to identify effective remediation strategies, including strategies to address the high rates of undiagnosed cases. Increasing psychiatric resources, general practitioner mental health competency, access to peer support, and increased efforts to reduce mental health stigma should be considered to address this growing public health crisis.

Deep learning attention-guided radiomics for COVID-19 chest radiograph classification.

Background: Accurate assessment of coronavirus disease 2019 (COVID-19) lung involvement through chest radiograph plays an important role in effective management of the infection. This study aims to develop a two-step feature merging method to integrate image features from deep learning and radiomics to differentiate COVID-19, non-COVID-19 pneumonia and normal chest radiographs (CXR). Methods: In this study, a deformable convolutional neural network (deformable CNN) was developed and used as a feature extractor to obtain 1,024-dimensional deep learning latent representation (DLR) features. Then 1,069-dimensional radiomics features were extracted from the region of interest (ROI) guided by deformable CNN's attention. The two feature sets were concatenated to generate a merged feature set for classification. For comparative experiments, the same process has been applied to the DLR-only feature set for verifying the effectiveness of feature concatenation. Results: Using the merged feature set resulted in an overall average accuracy of 91.0% for three-class classification, representing a statistically significant improvement of 0.6% compared to the DLR-only classification. The recall and precision of classification into the COVID-19 class were 0.926 and 0.976, respectively. The feature merging method was shown to significantly improve the classification performance as compared to using only deep learning features, regardless of choice of classifier (P value <0.0001). Three classes' F1-score were 0.892, 0.890, and 0.950 correspondingly (i.e., normal, non-COVID-19 pneumonia, COVID-19). Conclusions: A two-step COVID-19 classification framework integrating information from both DLR and radiomics features (guided by deep learning attention mechanism) has been developed. The proposed feature merging method has been shown to improve the performance of chest radiograph classification as compared to the case of using only deep learning features.

Home-based brain-computer interface attention training program for attention deficit hyperactivity disorder: a feasibility trial.

BACKGROUND: Attention deficit hyperactivity disorder (ADHD) is a prevalent child neurodevelopmental disorder that is treated in clinics and in schools. Previous trials suggested that our brain-computer interface (BCI)-based attention training program could improve ADHD symptoms. We have since developed a tablet version of the training program which can be paired with wireless EEG headsets. In this trial, we investigated the feasibility of delivering this tablet-based BCI intervention at home. METHODS: Twenty children diagnosed with ADHD, who did not receive any medication for the preceding month, were randomised to receive the 8-week tablet-based BCI intervention either in the clinic or at home. Those in the home intervention group received instructions before commencing the program and got reminders if they were lagging on the training sessions. The ADHD Rating Scale was completed by a blinded clinician at baseline and at week 8. Adverse events were monitored during any contact with the child throughout the trial and at week 8. RESULTS: Children in both groups could complete the tablet-based intervention easily on their own with minimal support from the clinic therapist or their parents (at home). The intervention was safe with few reported adverse effects. Clinician-rated inattentive symptoms on the ADHD-Rating Scale reduced by 3.2 (SD 6.20) and 3.9 (SD 5.08) for the home-based and clinic-based groups respectively, suggesting that home-based intervention was comparable to clinic-based intervention. CONCLUSIONS: This trial demonstrated that the tablet version of our BCI-based attention training program can be safely delivered to children in the comfort of their own home. Trial registration This trial is registered at clinicaltrials.gov as NCT01344044.

The role of anxiety and callous-unemotional traits in the relationship between externalizing behaviors and sleep problems in clinic-referred youth.

In a growing body of literature, poor sleep quality has been associated with externalizing problems. In adults, anxiety was found to mediate the relationship, and callous-unemotional (CU) traits were found to moderate it. We sought to examine these relationships in a child population. We examined these relationships in 239 clinic-referred youth (age 6-17) in Singapore with externalizing behavior problems. Parent- but not child-rated sleep problems were associated with increased parent-rated externalizing problems. This association was partially mediated by anxiety. Unlike in adults, CU traits did not moderate the relationship. Sleep problems were associated with externalizing problems regardless of the level of CU traits. It is possible externalizing behaviors may lead children to internalize experiences, leading to anxiety about their behaviors. Another possibility is externalizing behaviors may lead to more stressful life experiences due to negative reactions children with externalizing behaviors receive from parents, teachers, or peers. Regardless, the partial mediation found indicates anxiety may be an important factor to consider in future interventions focused on improving sleep as a means to reduce externalizing problems.

Development and validation of bone-suppressed deep learning classification of COVID-19 presentation in chest radiographs.

Background: Coronavirus disease 2019 (COVID-19) is a pandemic disease. Fast and accurate diagnosis of COVID-19 from chest radiography may enable more efficient allocation of scarce medical resources and hence improved patient outcomes. Deep learning classification of chest radiographs may be a plausible step towards this. We hypothesize that bone suppression of chest radiographs may improve the performance of deep learning classification of COVID-19 phenomena in chest radiographs. Methods: Two bone suppression methods (Gusarev et al. and Rajaraman et al.) were implemented. The Gusarev and Rajaraman methods were trained on 217 pairs of normal and bone-suppressed chest radiographs from the X-ray Bone Shadow Suppression dataset (https://www.kaggle.com/hmchuong/xray-bone-shadow-supression). Two classifier methods with different network architectures were implemented. Binary classifier models were trained on the public RICORD-1c and RSNA Pneumonia Challenge datasets. An external test dataset was created retrospectively from a set of 320 COVID-19 positive patients from Queen Elizabeth Hospital (Hong Kong, China) and a set of 518 non-COVID-19 patients from Pamela Youde Nethersole Eastern Hospital (Hong Kong, China), and used to evaluate the effect of bone suppression on classifier performance. Classification performance, quantified by sensitivity, specificity, negative predictive value (NPV), accuracy and area under the receiver operating curve (AUC), for non-suppressed radiographs was compared to that for bone suppressed radiographs. Some of the pre-trained models used in this study are published at (https://github.com/danielnflam). Results: Bone suppression of external test data was found to significantly (P<0.05) improve AUC for one classifier architecture [from 0.698 (non-suppressed) to 0.732 (Rajaraman-suppressed)]. For the other classifier architecture, suppression did not significantly (P>0.05) improve or worsen classifier performance. Conclusions: Rajaraman suppression significantly improved classification performance in one classification architecture, and did not significantly worsen classifier performance in the other classifier architecture. This research could be extended to explore the impact of bone suppression on classification of different lung pathologies, and the effect of other image enhancement techniques on classifier performance.

Association of cardiovascular metabolic risk factor measurements with psychiatric readmission among in-hospital patients with severe mental illness: a retrospective study.

BACKGROUND: Patients with severe mental illness (SMI) and comorbid physical conditions were often associated with higher risks of mortality and hospital readmission. In this study, we aim to examine the association of cardiovascular metabolic risk factor measurements with risks of psychiatric readmissions among in-hospital patients with severe mental illness (SMI). METHODS: We collected the longitudinal information of laboratory investigations, blood pressure and body mass index (BMI) among in-hospital patients who had been diagnosed with schizophrenia, major depression disorder or bipolar disorder and with comorbid diagnosis of hypertension, hyperlipidemia or diabetes from Jan 2014 to Jan 2019. The primary outcome was time to first psychiatric readmission. Cox proportional hazard model was utilized to calculate the hazard risks (HR) of cardiovascular metabolic risk factors with psychiatric readmission. RESULTS: A total of 5,256 patients were included in the analysis. Compared to patients with normal blood parameters, patients with aberrant tests of high-density dyslipidemia (HDL) and diastolic blood pressure (DBP) during in-hospitalization period were associated with higher risks to first psychiatric readmission [ HR (Hazard Ratio), 1.37 95% Confidence interval (CI), 1.03-1.83 for HDL and HR, 1.32 (95% CI, 1.04-1.67])for DBP]. Compared to patients with optimal monitoring, patients with suboptimal monitoring of blood lipids and blood pressure during in-hospitalization period or recommended window period of cardiovascular disease (CVD) risk management were associated with higher risks to first psychiatric readmission. CONCLUSIONS: Aberrant cardiovascular metabolic blood test and blood pressure and missing measurements among in-hospital patients with SMI were associated with increased risks of psychiatric readmissions. This calls for more active screening and monitoring of CVD risk factors for those in-hospital patients in need.

Feasibility trial of virtual reality exposure therapy for selective mutism.

Virtual reality exposure therapy (VRET) has been commonly utilised as an extension of cognitive behavioural therapy (CBT). However, most studies examined its effectiveness among adults, with no study focusing on children with selective mutism (SM). We aimed to examine its feasibility and acceptability among children with SM. Twenty children aged 6-12 with SM diagnosis were recruited and completed six therapist-guided VRET sessions. Parents and clinicians completed measures at pre-VRET, post-VRET, 1-month and 3-month follow-up visits. At post-VRET, parent and child participants completed the acceptability questionnaires. Findings suggested the feasibility of VRET as all participants completed the programme with no attrition. Parents and child participants also reported VRET to be an acceptable and effective treatment for SM. Significant improvement in overall functioning were found at post-treatment and follow-up measures, but there were no significant changes in parent-rated speech frequency and anxiety measures. These support the acceptability of VRET as an adjunct modality (and not substitute) of CBT in SM treatment. Future studies, with more robust experimental designs and larger sample sizes, can be conducted to confirm its efficacy. As technology becomes more sophisticated, tools such as virtual environments can be explored to enhance evidence-based care for children and their families.

The making of child and adolescent psychiatrists in Australia and New Zealand.

Child and adolescent psychiatry has an opportunity to adapt alongside the advancements in medical knowledge, post-graduate training, epidemiological realities and clinical service models. Here, we are guided by the mental health needs of our community's young and their families in our review of child and adolescent psychiatry training in Australia and New Zealand. We recognise that training must respond to clinical demand and service reform while ensuring a range of clinical and educational experiences to deliver expectable competencies in order to produce child and adolescent psychiatrists that meet the communities' needs now and in the future. We argue that training programmes be subject to rigorous evaluation by embedding continuing cycles of improvement including regular review and international bench marking.

Differences in the Prevalence and Profile of DSM-IV and DSM-5 Alcohol Use Disorders-Results from the Singapore Mental Health Study 2016.

INTRODUCTION: The Diagnostic and Statistical Manual of Mental Disorders, 5th edition (DSM-5) criteria for alcohol use disorders (AUD) was a significant shift from the historical DSM-IV Text Revised version. Following this shift in diagnostic criteria, a difference in the prevalence of AUD was expected. The current study aimed to evaluate the consequences of the modification of the diagnostic criteria from DSM-IV to DSM-5 AUD using lifetime diagnosis in Singapore's multi-ethnic population using data from a nationwide epidemiological study. METHODS: Respondents were assessed for lifetime mental disorders using the Composite International Diagnostic Interview (CIDI) administered through face-to-face interviews. Lifetime DSM-IV AUD diagnoses were compared with DSM-5 AUD diagnoses generated by modifying the criteria and the addition of the craving criterion. Sociodemographic correlates of lifetime DSM-IV/DSM-5 AUD were examined using multiple logistic regression analysis. Associations between DSM-IV/DSM-5 AUD, chronic conditions, and the HRQOL summary scores were examined using logistic and linear regression after controlling for significant sociodemographic factors. RESULTS: The lifetime prevalence of DSM-IV AUD was 4.6% (0.5% for dependence and 4.1% for abuse) in the adult population, while the lifetime prevalence of DSM-5 AUD was 2.2%. Younger age, male gender, and lower education were associated with higher odds of both DSM-IV and DSM-5 AUD. However, those who were economically inactive (versus employed) (AOR, 0.4; 95% CI 0.2-0.9) and had a higher monthly household income (SGD 4000-5999 versus below SGD 2000) had lower odds of DSM-IV AUD (AOR, 0.4; 95% CI 0.2-0.7), but this was not observed among those with DSM-5 AUD. Both DSM-IV and DSM-5 AUD were associated with significant comorbidities in terms of other mental disorders; however, those diagnosed with lifetime GAD had significantly higher odds of having DSM-5 AUD (AOR, 5; 95% CI 1.9-13.2) but not DSM-IV AUD. CONCLUSIONS: While using the DSM-5 criteria would result in a lower prevalence of AUD in Singapore, it remains a highly comorbid condition associated with a poor health-related quality of life that is largely untreated, which makes it a significant public health concern.

Automated classification of attention deficit hyperactivity disorder and conduct disorder using entropy features with ECG signals.

BACKGROUND: The most prevalent neuropsychiatric disorder among children is attention deficit hyperactivity disorder (ADHD). ADHD presents with a high prevalence of comorbid disorders such as conduct disorder (CD). The lack of definitive confirmatory diagnostic tests for ADHD and CD make diagnosis challenging. The distinction between ADHD, ADHD + CD and CD is important as the course and treatment are different. Electrocardiography (ECG) signals may become altered in behavioral disorders due to brain-heart autonomic interactions. We have developed a software tool to categorize ADHD, ADHD + CD and CD automatically on ECG signals. METHOD: ECG signals from participants were decomposed using empirical wavelet transform into various modes, from which entropy features were extracted. Robust ten-fold cross-validation with adaptive synthetic sampling (ADASYN) and z-score normalization were performed at each fold. Analysis of variance (ANOVA) technique was employed to determine the variability within the three classes, and obtained the most discriminatory features. Highly significant entropy features were then fed to classifiers. RESULTS: Our model yielded the best classification results with the bagged tree classifier: 87.19%, 87.71% and 86.29% for accuracy, sensitivity and specificity, respectively. CONCLUSION: The proposed expert system can potentially assist mental health professionals in the stratification of the three classes, for appropriate intervention using accessible ECG signals.

Validity of the patient health questionnaire 9-item in autistic youths: a pilot study.

BACKGROUND: Autistic adolescents have greater predisposition to depression and suicidality than neurotypical adolescents. Early detection is essential for timely treatment. The Patient Health Questionnaire 9-item (PHQ-9) is a brief screen for depression. The study examines the validity of the PHQ-9 for detecting major depressive disorder (MDD) in autistic youths. METHODS: English speaking youths aged 10-18 years, with DSM-IV/DSM-5/ICD-10 diagnosis of Autism Spectrum Disorder (ASD), and their parents presenting to a child psychiatric service were invited to participate between May 2018 to August 2020. Participants completed the respective self- and parent-rated PHQ-9 independently. MDD was verified using the MINI-Kid (Mini-International Neuropsychiatric Interview, Kid version). RESULTS: One hundred one youth, mean (SD) age 14.6 (2.3), were enrolled. 27 (27%) met criteria for current MDD. Mean total PHQ-9 scores, percentage ratings for severity of symptoms of depression, functional impairment, dysthymia and suicidality were compared. Areas under the ROC curve and statistically optimal cutoffs were determined. Parents rated depressive symptoms severity lower than their children. The PHQ-9 displayed low sensitivity with high false negative rates at conventional, adjusted and proposed cutoffs. CONCLUSIONS: Future studies should improve on the validity and reliability of existing depression screening tools, or develop more appropriate screening methods of depression, for autistic youths.

Mental Resilience of Medical Practitioners in Singapore during COVID-19: Survey Results from a Webinar Course on Resilience.

Front-line doctors are at high risk of exposure to COVID-19. The mental resilience of general practitioners and their areas of concerns and support required are important during this COVID-19 period. A total of 403 general practitioners attending a webinar on resiliency, hosted by the College of Family Physicians, Singapore, participated in the survey anonymously. Participants provided responses to questions relating to COVID-19 in the domains of Family and friends, Myself, Practice, and Community. Responses are categorized into LEARNING, FEAR, and GROWTH zones. The majority of the doctors reported to be in the GROWTH zone in relation to Family and friends (39%) and Myself (38%) as compared to Practice and Community, aOR = 4.5 (95% CI 3.4 to 5.9), p < 0.001. 34% of the participants reported being in the FEAR zone in relation to Family and friends, aOR = 8.0 (95% CI = 5.6 to 11.2), p < 0.001; at least 81% reported being in the LEARNING zone in relation to Practice and Community, aOR = 7.5 (95% CI = 5.8 to 9.6), p < 0.001, compared to other domains. Supporting and protecting the doctors is important in strategic planning and management of the current pandemic and building preparedness and an effective response towards future crises.

Our children then and now: Changes in mental health symptoms among Singaporean children from 2003 to 2017.

In recent years, there appears to be a rise in the diagnosis and treatment of child mental health disorders in many countries, including Singapore. While this increase may be alarming, it could possibly be attributed to factors such as changes in diagnostic criteria, improved screening in schools and primary health settings, changes in clinical practices, and an increase in help-seeking behaviour. Hence, an examination of community-level trends in mental health symptoms can elucidate how child psychopathology has changed over the years. This study aimed to investigate differences in symptoms of mental health between two cohorts of young Asian children aged six to eight living in Singapore. Child Behaviour Checklist (CBCL) scores from a sample in 2003 (Cohort 1; n = 524) were compared to another sample taken in 2017 (Cohort 2; n = 655). Cohort 2 had lower externalizing scale scores as compared to Cohort 1, but there were no significant differences in total problem scores or internalizing scale scores. Among the CBCL subscales, Cohort 2 had comparatively lower levels of aggressive behaviour and withdrawn/depressed symptoms, but higher levels of thought problems and somatic complaints as compared to Cohort 1. Our findings suggest that children in Singapore are progressing as well as, or even better than, children 14 years ago on most aspects of mental well-being.

Comparison of mental-physical comorbidity, risk of death and mortality among patients with mental disorders - A retrospective cohort study.

AIM: To compare the risk of death, the prevalence of comorbid chronic physical illness and mortality among an Asian population of patients with mental disorders. METHODS: This was a retrospective data analysing of medical records of patients with schizophrenia, depression, anxiety, bipolar disorder, alcohol use disorder (AUD) or substance use disorder and the comorbid chronic physical illnesses. The hazard risk of death was calculated with Cox regression and compared between patients with and without comorbid chronic physical illness(es). Odds ratios of specific comorbid chronic physical illness were calculated with logistic regression and mean crude death rate was calculated for patients with different mental disorders. RESULTS: A total of 56,447 patients with mental disorders were included in the analysis. Compared to patients without comorbid physical illness, patients with mental-physical comorbidity were associated with a higher risk of death [2.36 (2.22-2.52); hazard ratio (95% CI)] and less estimated survival days [2157 (2142-2172) vs 2508 (2504-2513)]. Compared to other mental disorders, those with AUD had the highest prevalence of two or more comorbid chronic physical illnesses and associated with the highest odds of comorbid hypertension, diabetes mellitus, stroke, nephritis, chronic kidney disease, and cancer. The highest one-year crude death rate was similarly observed in patients with AUD. CONCLUSIONS: Mental-physical comorbidity was associated with a higher risk of death compared to patients with mental disorders only. The highest prevalence of mental-physical comorbidity and mortality were observed in patients with AUD. More attention and resources may be needed to tackle the burden of AUD.

Quantitative prediction of the extent of pelvic tumour ablation by magnetic resonance-guided high intensity focused ultrasound.

BACKGROUND: Patient suitability for magnetic resonance-guided high intensity focused ultrasound (MRgHIFU) therapy of pelvic tumors is currently assessed by visual estimation of the proportion of tumor that can be reached by the device's focus (coverage). Since it is important to assess whether enough energy reaches the tumor to achieve ablation, a methodology for estimating the proportion of the tumor that can be ablated (treatability) was developed. Predicted treatability was compared against clinically achieved thermal ablation. METHODS: MR Dixon sequence images of five patients with recurrent gynecological tumors were acquired during their treatment. Acousto-thermal simulations were performed using k-Wave for three exposure points (the deepest and shallowest reachable focal points within the tumor, identified from tumor coverage analysis, and a point halfway in-between) per patient. Interpolation between the resulting simulated ablated tissue volumes was used to estimate the maximum treatable depth and hence, tumor treatability. Predicted treatability was compared both to predicted tumor coverage and to the clinically treated tumor volume. The intended and simulated volumes and positions of ablated tissues were compared. RESULTS: Predicted treatability was less than coverage by 52% (range: 31-78%) of the tumor volume. Predicted and clinical treatability differed by 9% (range: 1-25%) of tumor volume. Ablated tissue volume and position varied with beam path length through tissue. CONCLUSION: Tumor coverage overestimated patient suitability for MRgHIFU therapy. Employing patient-specific simulations improved treatability assessment. Patient treatability assessment using simulations is feasible.