NIMG-28. PROSPECTIVE HISTOPATHOLOGY-VALIDATED MACHINE LEARNING FOR DISTINGUISHING TRUE PROGRESSION FROM TREATMENT-RELATED CHANGES IN GLIOBLASTOMA PATIENTS

2021 ◽  
Vol 23 (Supplement_6) ◽  
pp. vi134-vi135
Author(s):  
Hamed Akbari ◽  
Anahita Fathi Kazerooni ◽  
Spyridon Bakas ◽  
Chiharu Sako ◽  
Elizabeth Mamourian ◽  
...  
Keyword(s):  
Machine Learning ◽  
Tumor Resection ◽  
Water Concentration ◽  
Dystrophic Calcification ◽  
Non Invasive ◽  
Mixed Response ◽  
Second Resection ◽  
Resected Tissue ◽  
Mitotic Figures

Abstract PURPOSE Decision making about the best course of treatment for glioblastoma patients becomes challenging when a new enhancing lesion appears in the vicinity of the surgical bed on follow-up MRI (after maximal safe tumor resection and chemoradiation), raising concerns for tumor progression (TP). Literature indicates 30-50% of these new lesions describe primarily treatment-related changes (TRC). We hypothesize that quantitative analysis of specific and sensitive features extracted from multi-parametric MRI (mpMRI) via machine learning (ML) techniques may yield non-invasive imaging signatures that distinguish TP from TRC and facilitate better treatment personalization. METHODS We have generated an ML model on a retrospective cohort of 58 subjects, and prospectively evaluated on an independent cohort of 58 previously unseen patients who underwent second resection for suspicious recurrence and had availability of advanced mpMRI (T1, T1-Gd, T2, T2-FLAIR, DTI, DSC). The features selected by our retrospective model, representing principal components analysis of intensity distributions, morphological, statistical, and texture descriptors, were extracted from the mpMRI of the prospective cohort. Integration of these features revealed signatures distinguishing between TP, mixed response, and TRC. Independently, a board-certified neuropathologist evaluated the resected tissue by blindly classifying it in the above three categories, based on mitotic figures, pseudopalisading necrosis, geographic necrosis, dystrophic calcification, vascular changes, and Ki67. RESULTS Tissues classified as TRC by the neuropathologist were associated with imaging phenotypes of lower angiogenesis (DSC-derived features), lower cellularity (DTI-derived features), and higher water concentration (T2, T2-FLAIR features). Our ML model characterized TP with 78% accuracy (sensitivity:86%, specificity:70%, AUC:0.80 (95%CI, 0.68-0.92)) and TRC with 81% accuracy (sensitivity:80%, specificity:81%, AUC:0.87 (95%CI, 0.72-1.00)). CONCLUSION Our proposed ML model reveals distinct non-invasive markers of TP and TRC, directly associated with histopathological changes in prospective glioblastoma patients. Reliable stratification of TP and TRC entities may help to noninvasively determine whether the course of treatment should change.

scholarly journals Machine Learning Based Analysis of Human Serum N-glycome Alterations to Follow up Lung Tumor Surgery

Cancers ◽  
2020 ◽  
Vol 12 (12) ◽  
pp. 3700
Author(s):  
Brigitta Mészáros ◽  
Gábor Járvás ◽  
Renáta Kun ◽  
Miklós Szabó ◽  
Eszter Csánky ◽  
...  
Keyword(s):  
Machine Learning ◽  
Data Analysis ◽  
Human Serum ◽  
Serum Proteins ◽  
Lung Tumor ◽  
Classification Tree ◽  
Tumor Resection ◽  
Fluorescent Detection ◽  
Glycosylation Pattern

The human serum N-glycome is a valuable source of biomarkers for malignant diseases, already utilized in multiple studies. In this paper, the N-glycosylation changes in human serum proteins were analyzed after surgical lung tumor resection. Seventeen lung cancer patients were involved in this study and the N-glycosylation pattern of their serum samples was analyzed before and after the surgery using capillary electrophoresis separation with laser-induced fluorescent detection. The relative peak areas of 21 N-glycans were evaluated from the acquired electropherograms using machine learning-based data analysis. Individual glycans as well as their subclasses were taken into account during the course of evaluation. For the data analysis, both discrete (e.g., smoker or not) and continuous (e.g., age of the patient) clinical parameters were compared against the alterations in these 21 N-linked carbohydrate structures. The classification tree analysis resulted in a panel of N-glycans, which could be used to follow up on the effects of lung tumor surgical resection.

Bladder Wall Calcification Following Transurethral Resection of Bladder Tumor and Intravesical Mitomycin C Instillation

2018 ◽  
Vol 5 ◽  
pp. 3-5
Author(s):  
Sarah K. Perz ◽  
Patrick M Tenbrink ◽  
Gregory H. Haselhuhn
Keyword(s):  
Mitomycin C ◽  
Transurethral Resection ◽  
Bladder Wall ◽  
Tumor Resection ◽  
Dystrophic Calcification ◽  
Non Invasive ◽  
Voiding Symptoms ◽  
Wall Calcification ◽  
Added Benefit ◽  
Carcinoma Of The Bladder

In rarely reported instances, the use of intravesical mitomycin C appears to have caused bladder wall calcification. We report two patients treated for non-invasive urothelial carcinoma of the bladder with transurethral resection and mitomycin c instillation. Both of these patients experienced new onset of severe irritative voiding symptoms shortly thereafter. Although one patient com- pleted a six-cycle course of Bacillus Calmette-Guerin (BCG), the other only received one instillation due to symptom severity. On cystoscopy the resection beds appeared calcified and necrotic. Biopsies revealed dystrophic calcification and necrosis without evidence of tumor. Resection of this area led to complete resolu- tion of symptoms and normal healing of the resection site. Our results are consistent with the current hypothesis that these cal- cifications are not related to recurrence or persistence of tumor. Although there have not been reported cases of these calcifica- tions signifying recurrence of tumor, it is important for clinicians to rule out recurrence as a cause for these symptoms. Resection of the area has the added benefit of alleviation of irritative void- ing symptoms encountered shortly after mitomycin c instillation.

scholarly journals New risk model is able to identify patients with a low risk of progression in systemic sclerosis

RMD Open ◽  
2021 ◽  
Vol 7 (2) ◽  
pp. e001524
Author(s):  
Nina Marijn van Leeuwen ◽  
Marc Maurits ◽  
Sophie Liem ◽  
Jacopo Ciaffi ◽  
Nina Ajmone Marsan ◽  
...  
Keyword(s):  
Machine Learning ◽  
Systemic Sclerosis ◽  
High Risk ◽  
Disease Progression ◽  
Risk Model ◽  
Immunosuppressive Drugs ◽  
Low Risk ◽  
Expert Assessment ◽  
Risk Of Progression

ObjectivesTo develop a prediction model to guide annual assessment of systemic sclerosis (SSc) patients tailored in accordance to disease activity.MethodsA machine learning approach was used to develop a model that can identify patients without disease progression. SSc patients included in the prospective Leiden SSc cohort and fulfilling the ACR/EULAR 2013 criteria were included. Disease progression was defined as progression in ≥1 organ system, and/or start of immunosuppression or death. Using elastic-net-regularisation, and including 90 independent clinical variables (100% complete), we trained the model on 75% and validated it on 25% of the patients, optimising on negative predictive value (NPV) to minimise the likelihood of missing progression. Probability cutoffs were identified for low and high risk for disease progression by expert assessment.ResultsOf the 492 SSc patients (follow-up range: 2–10 years), disease progression during follow-up was observed in 52% (median time 4.9 years). Performance of the model in the test set showed an AUC-ROC of 0.66. Probability score cutoffs were defined: low risk for disease progression (<0.197, NPV:1.0; 29% of patients), intermediate risk (0.197–0.223, NPV:0.82; 27%) and high risk (>0.223, NPV:0.78; 44%). The relevant variables for the model were: previous use of cyclophosphamide or corticosteroids, start with immunosuppressive drugs, previous gastrointestinal progression, previous cardiovascular event, pulmonary arterial hypertension, modified Rodnan Skin Score, creatine kinase and diffusing capacity for carbon monoxide.ConclusionOur machine-learning-assisted model for progression enabled us to classify 29% of SSc patients as ‘low risk’. In this group, annual assessment programmes could be less extensive than indicated by international guidelines.

Primary Calvarial Ewing Sarcoma: A Case Series

2021 ◽  
Author(s):  
Sandeep Mohindra ◽  
Manjul Tripathi ◽  
Aman Batish ◽  
Ankur Kapoor ◽  
Ninad Ramesh Patil ◽  
...  
Keyword(s):  
Ewing Sarcoma ◽  
Tumor Resection ◽  
Case Series ◽  
Radical Excision ◽  
Term Outcome ◽  
Long Term Outcome ◽  
Intradural Tumor ◽  
Age Range

Abstract Background Calvarial Ewing tumor is a relatively rare differential among bony neoplasms. We present our experience of managing primary calvarial Ewing sarcoma (EWS), highlighting their clinical and radiological findings. Method In a retrospective analysis, we evaluated our 12-year database for pathologically proven EWS. A literature search was conducted for the comparative presentation and update on the management and outcome. Result From January 2008 to December 2020, we managed eight patients (male:female = 5:3; age range 6 months to 19 years, mean 11.5 years) harboring primary calvarial EWS. All cases underwent wide local excision; two patients required intradural tumor resection, while one required rotation flap for scalp reconstruction. Mean hospital stay was 8 days. All patients received adjuvant chemo- and radiotherapy. Three patients remained asymptomatic at 5 years of follow-up, while two patients died. Conclusion Primary calvarial EWS is a rare entity. It usually affects patients in the first two decades of life. These tumors can be purely intracranial, causing raised intracranial pressure symptoms, which may exhibit rapidly enlarging subgaleal tumors with only cosmetic deformities or symptoms of both. Radical excision followed by adjuvant therapy may offer a favorable long-term outcome.

scholarly journals Taxonomic signatures of cause-specific mortality risk in human gut microbiome

2021 ◽  
Vol 12 (1) ◽  
Author(s):  
Aaro Salosensaari ◽  
Ville Laitinen ◽  
Aki S. Havulinna ◽  
Guillaume Meric ◽  
Susan Cheng ◽  
...  
Keyword(s):  
Mortality Risk ◽  
Gut Microbiome ◽  
Human Gut ◽  
Cross Sectional ◽  
Microbiome Composition ◽  
Human Gut Microbiome ◽  
Non Invasive ◽  
Sequencing Technologies

AbstractThe collection of fecal material and developments in sequencing technologies have enabled standardised and non-invasive gut microbiome profiling. Microbiome composition from several large cohorts have been cross-sectionally linked to various lifestyle factors and diseases. In spite of these advances, prospective associations between microbiome composition and health have remained uncharacterised due to the lack of sufficiently large and representative population cohorts with comprehensive follow-up data. Here, we analyse the long-term association between gut microbiome variation and mortality in a well-phenotyped and representative population cohort from Finland (n = 7211). We report robust taxonomic and functional microbiome signatures related to the Enterobacteriaceae family that are associated with mortality risk during a 15-year follow-up. Our results extend previous cross-sectional studies, and help to establish the basis for examining long-term associations between human gut microbiome composition, incident outcomes, and general health status.

scholarly journals Non-invasive cuff-less blood pressure estimation using a hybrid deep learning model

2021 ◽  
Vol 53 (2) ◽  
Author(s):  
Sen Yang ◽  
Yaping Zhang ◽  
Siu-Yeung Cho ◽  
Ricardo Correia ◽  
Stephen P. Morgan
Keyword(s):  
Machine Learning ◽  
Blood Pressure ◽  
Deep Learning ◽  
Absolute Error ◽  
Learning Model ◽  
Physiological Measurement ◽  
British Hypertension Society ◽  
Non Invasive ◽  
Blood Pressure Estimation ◽  
Deep Learning Model

AbstractConventional blood pressure (BP) measurement methods have different drawbacks such as being invasive, cuff-based or requiring manual operations. There is significant interest in the development of non-invasive, cuff-less and continual BP measurement based on physiological measurement. However, in these methods, extracting features from signals is challenging in the presence of noise or signal distortion. When using machine learning, errors in feature extraction result in errors in BP estimation, therefore, this study explores the use of raw signals as a direct input to a deep learning model. To enable comparison with the traditional machine learning models which use features from the photoplethysmogram and electrocardiogram, a hybrid deep learning model that utilises both raw signals and physical characteristics (age, height, weight and gender) is developed. This hybrid model performs best in terms of both diastolic BP (DBP) and systolic BP (SBP) with the mean absolute error being 3.23 ± 4.75 mmHg and 4.43 ± 6.09 mmHg respectively. DBP and SBP meet the Grade A and Grade B performance requirements of the British Hypertension Society respectively.

scholarly journals Non-Invasive Dengue Diagnostics—The Use of Saliva and Urine for Different Stages of the Illness

Diagnostics ◽  
2021 ◽  
Vol 11 (8) ◽  
pp. 1345
Author(s):  
Mahathir Humaidi ◽  
Wei Ping Tien ◽  
Grace Yap ◽  
Choon Rong Chua ◽  
Lee Ching Ng
Keyword(s):  
Clinical Symptoms ◽  
Detection Algorithm ◽  
Acute Stage ◽  
Serum Samples ◽  
Non Invasive ◽  
Fever Onset ◽  
Dengue Transmission ◽  
Dengue Diagnostics ◽  
Potential Tool

Dengue diagnosis is largely dependent on clinical symptoms and routinely confirmed with laboratory detection of dengue virus in patient serum samples collected via phlebotomy. This presents a challenge to patients not amenable to venipuncture. Non-invasive methods of dengue diagnosis have the potential to enhance the current dengue detection algorithm. In this study, samples from dengue infected patients were collected between January 2012 until September 2012 and September 2013 until December 2013 in two different setups. Panel A samples (blood, urine, and saliva) were collected daily when the 39 patients were hospitalised and during their follow-up visits while Panel B samples (saliva) were collected from 23 patients during the acute stage of dengue. Using DENV PCR on Panel A, from day 2 to day 4 post fever onset, serum showed the best overall positivity followed by saliva and urine (100%/82.1%/67.9%). From day 5 until day 10 post fever onset, serum and urine had similar positivity (67.4%/61.2%), followed by saliva (51.3%). Beyond day 10 post fever onset, DENV was undetectable in sera, but urine and saliva showed 56.8% and 28.6% positivity, respectively. DENV in urine was detectable up until 32 days post fever. Panel B results showed overall sensitivity of 32.4%/36% (RNA/NS1) for DENV detection in saliva. Our results suggest that the urine-based detection method is useful especially for late dengue detection, where DENV is undetected in sera but still detectable in urine. This provides a potential tool for the physician to pick up new cases in an area where there is ongoing dengue transmission and subsequently prompt for intensified vector control activities.

scholarly journals RONC-19. TWO CASES OF RE-IRRADIATION FOR LATE RECURRENT OR RADIATION-INDUCED TUMOR AFTER RADIATION THERAPY FOR PEDIATRIC BRAIN TUMORS

2020 ◽  
Vol 22 (Supplement_3) ◽  
pp. iii459-iii459
Author(s):  
Takashi Mori ◽  
Shigeru Yamaguchi ◽  
Rikiya Onimaru ◽  
Takayuki Hashimoto ◽  
Hidefumi Aoyama
Keyword(s):  
Radiation Therapy ◽  
Brain Tumors ◽  
Tumor Resection ◽  
Intensity Modulated Radiation Therapy ◽  
Late Recurrence ◽  
Pediatric Brain Tumors ◽  
Suprasellar Region ◽  
Radiation Induced ◽  
Pediatric Brain

Abstract BACKGROUND As the outcome of pediatric brain tumors improves, late recurrence and radiation-induced tumor cases are more likely to occur, and the number of cases requiring re-irradiation is expected to increase. Here we report two cases performed intracranial re-irradiation after radiotherapy for pediatric brain tumors. CASE 1: 21-year-old male. He was diagnosed with craniopharyngioma at eight years old and underwent a tumor resection. At 10 years old, the local recurrence of suprasellar region was treated with 50.4 Gy/28 fr of stereotactic radiotherapy (SRT). After that, other recurrent lesions appeared in the left cerebellopontine angle, and he received surgery three times. The tumor was gross totally resected and re-irradiation with 40 Gy/20 fr of SRT was performed. We have found no recurrence or late effects during the one year follow-up. CASE 2: 15-year-old female. At three years old, she received 18 Gy/10 fr of craniospinal irradiation and 36 Gy/20 fr of boost to the posterior fossa as postoperative irradiation for anaplastic ependymoma and cured. However, a anaplastic meningioma appeared on the left side of the skull base at the age of 15, and 50 Gy/25 fr of postoperative intensity-modulated radiation therapy was performed. Two years later, another meningioma developed in the right cerebellar tent, and 54 Gy/27 fr of SRT was performed. Thirty-three months after re-irradiation, MRI showed a slight increase of the lesion, but no late toxicities are observed. CONCLUSION The follow-up periods are short, however intracranial re-irradiation after radiotherapy for pediatric brain tumors were feasible and effective.

scholarly journals Risk assessment in precapillary pulmonary hypertension: a comparative analysis

2021 ◽  
Vol 22 (1) ◽  
Author(s):  
Thomas Sonnweber ◽  
Eva-Maria Schneider ◽  
Manfred Nairz ◽  
Igor Theurl ◽  
Günter Weiss ◽  
...  
Keyword(s):  
Risk Assessment ◽  
Pulmonary Hypertension ◽  
High Risk ◽  
Risk Stratification ◽  
Risk Prediction ◽  
Mortality Risk ◽  
Assessment Tool ◽  
Risk Models ◽  
Non Invasive

Abstract Background Risk stratification is essential to assess mortality risk and guide treatment in patients with precapillary pulmonary hypertension (PH). We herein compared the accuracy of different currently used PH risk stratification tools and evaluated the significance of particular risk parameters. Methods We conducted a retrospective longitudinal observational cohort study evaluating seven different risk assessment approaches according to the current PH guidelines. A comprehensive assessment including multi-parametric risk stratification was performed at baseline and 4 yearly follow-up time-points. Multi-step Cox hazard analysis was used to analyse and refine risk prediction. Results Various available risk models effectively predicted mortality in patients with precapillary pulmonary hypertension. Right-heart catheter parameters were not essential for risk prediction. Contrary, non-invasive follow-up re-evaluations significantly improved the accuracy of risk estimations. A lack of accuracy of various risk models was found in the intermediate- and high-risk classes. For these patients, an additional evaluation step including assessment of age and right atrium area improved risk prediction significantly. Discussion Currently used abbreviated versions of the ESC/ERS risk assessment tool, as well as the REVEAL 2.0 and REVEAL Lite 2 based risk stratification, lack accuracy to predict mortality in intermediate- and high-risk precapillary pulmonary hypertension patients. An expanded non-invasive evaluation improves mortality risk prediction in these individuals.

Sign in / Sign up

Export Citation Format

Share Document