Machine Learning-Based Ultrasound Texture Analysis in Differentiation of Benign Phyllodes Tumors from Borderline-Malignant Phyllodes Tumors

Author(s):  
Isil Basara Akin ◽  
Hakan Abdullah Ozgul ◽  
Canan Altay ◽  
Merih Guray Durak ◽  
Suleyman Ozkan Aksoy ◽  
...  

Abstract Purpose Phyllodes tumors (PTs) are uncommon fibroepithelial breast lesions that are classified as three different forms as benign phyllodes tumor (BPT), borderline phyllodes tumor (BoPT), and malignant phyllodes tumor (MPT). Conventional radiologic methods make only a limited contribution to exact diagnosis, and texture analysis data increase the diagnostic performance. In this study, we aimed to evaluate the contribution of texture analysis of US images (TAUI) of PTs in order to discriminate between BPTs and BoPTs-MPTs. Methods The number of patients was 63 (41 BPTs, 12 BoPTs, and 10 MPTs). Patients were divided into two groups (Group 1-BPT, Group 2-BoPT/MPT). TAUI with LIFEx software was performed retrospectively. An independent machine learning approach, MATLAB R2020a (Math- Works, Natick, Massachusetts) was used with the dataset with p < 0.004. Two machine learning approaches were used to build prediction models for differentiating between Group 1 and Group 2. Receiver operating characteristics (ROC) curve analyses were performed to evaluate the diagnostic performance of statistically significant texture data between phyllodes subgroups. Results In TAUI, 10 statistically significant second order texture values were identified as significant factors capable of differentiating among the two groups (p < 0.05). Both of the models of our dataset make a diagnostic contribution to the discrimination between BopTs-MPTs and BPTs. Conclusion In PTs, US is the main diagnostic method. Adding machine learning-based TAUI to conventional US findings can provide optimal diagnosis, thereby helping to choose the correct surgical method. Consequently, decreased local recurrence rates can be achieved.

2020 ◽  
Vol 22 (Supplement_2) ◽  
pp. ii158-ii158
Author(s):  
Nicholas Nuechterlein ◽  
Beibin Li ◽  
James Fink ◽  
David Haynor ◽  
Eric Holland ◽  
...  

Abstract BACKGROUND Previously, we have shown that combined whole-exome sequencing (WES) and genome-wide somatic copy number alteration (SCNA) information can separate IDH1/2-wildtype glioblastoma into two prognostic molecular subtypes (Group 1 and Group 2) and that these subtypes cannot be distinguished by epigenetic or clinical features. However, the potential for radiographic features to discriminate between these molecular subtypes has not been established. METHODS Radiogenomic features (n=35,400) were extracted from 46 multiparametric, pre-operative magnetic resonance imaging (MRI) of IDH1/2-wildtype glioblastoma patients from The Cancer Imaging Archive, all of whom have corresponding WES and SCNA data in The Cancer Genome Atlas. We developed a novel feature selection method that leverages the structure of extracted radiogenomic MRI features to mitigate the dimensionality challenge posed by the disparity between the number of features and patients in our cohort. Seven traditional machine learning classifiers were trained to distinguish Group 1 versus Group 2 using our feature selection method. Our feature selection was compared to lasso feature selection, recursive feature elimination, and variance thresholding. RESULTS We are able to classify Group 1 versus Group 2 glioblastomas with a cross-validated area under the curve (AUC) score of 0.82 using ridge logistic regression and our proposed feature selection method, which reduces the size of our feature set from 35,400 to 288. An interrogation of the selected features suggests that features describing contours in the T2 abnormality region on the FLAIR MRI modality may best distinguish these two groups from one another. CONCLUSIONS We successfully trained a machine learning model that allows for relevant targeted feature extraction from standard MRI to accurately predict molecularly-defined risk-stratifying IDH1/2-wildtype glioblastoma patient groups. This algorithm may be applied to future prospective studies to assess the utility of MRI as a surrogate for costly prognostic genomic studies.


Angiology ◽  
2021 ◽  
pp. 000331972199141
Author(s):  
Arafat Yildirim ◽  
Mehmet Kucukosmanoglu ◽  
Fethi Yavuz ◽  
Nermin Yildiz Koyunsever ◽  
Yusuf Cekici ◽  
...  

Many parameters included in the Anticoagulation and Risk Factors in Atrial Fibrillation (ATRIA) and CHA2DS2-VASc (congestive heart failure, hypertension, age ≥75 years, diabetes mellitus, stroke, vascular disease, age 65-74 years, sex category) scores also predict coronary artery disease (CAD). We modified the ATRIA score (ATRIA-HSV) by adding hyperlipidemia, smoking, and vascular disease and also male sex instead of female. We evaluated whether the CHA2DS2-VASc, CHA2DS2-VASc-HS, ATRIA, and ATRIA-HSV scores predict severe CAD. Consecutive patients with coronary angiography were prospectively included. A ≥50% stenosis in ≥1epicardial coronary artery (CA) was defined as severe CAD. Patient with normal CA (n = 210) were defined as group 1, with <50% CA stenosis (n = 178) as group 2, and with ≥50% stenosis (n = 297) as group 3. The mean ATRIA, ATRIA-HSV, CHA2DS2-VASc, and CHA2DS2VASc-HS scores increased from group 1 to group 3. A correlation was found between the Synergy between PCI with Taxus and Cardiac Surgery score and ATRIA ( r = 0.570), ATRIA-HSV ( r = 0.614), CHA2DS2-VASc ( r = 0.428), and CHA2DS2-VASc-HS ( r = 0.500) scores ( Ps < .005). Pairwise comparisons of receiver operating characteristics curves showed that ATRIA-HSV (>3 area under curve [AUC]: 0.874) and ATRIA (>3, AUC: 0.854) have a better performance than CHA2DS2-VASc (>1, AUC: 0.746) and CHA2DS2-VASc-HS (>2, AUC: 0.769). In conclusion, the ATRIA and ATRIA-HSV scores are simple and may be useful to predict severe CAD.


2021 ◽  
Vol 12 ◽  
pp. 215145932110096
Author(s):  
Christina Polan ◽  
Heinz-Lothar Meyer ◽  
Manuel Burggraf ◽  
Monika Herten ◽  
Paula Beck ◽  
...  

Background: The COVID-19 pandemic is challenging healthcare systems worldwide. This study examines geriatric patients with proximal femur fractures during the COVID-19 pandemic, shifts in secondary disease profile, the impact of the pandemic on hospitalization and further treatment. Methods: In a retrospective monocentric study, geriatric proximal femur fractures treated in the first six months of 2020 were analyzed and compared with the same period of 2019. Pre-traumatic status (living in a care home, under supervision of a legal guardian), type of trauma, accident mechanism, geriatric risk factors, associated comorbidities, time between hospitalization and surgery, inpatient time and post-operative further treatment of 2 groups of patients, aged 65-80 years (Group 1) and 80+ years (Group 2) were investigated. Results: The total number of patients decreased (70 in 2019 vs. 58 in 2020), mostly in Group 1 (25 vs. 16) while the numbers in Group 2 remained almost constant (45 vs. 42). The percentage of patients with pre-existing neurological conditions rose in 2020. This corresponded to an increase in patients under legal supervision (29.3%) and receiving pre-traumatic care in a nursing home (14.7%). Fractures were mostly caused by minor trauma in a home environment. In 2020, total number of inpatient days for Group 2 was lower compared to Group 1 (p = 0.008). Further care differed between the years: fewer Group 1 patients were discharged to geriatric therapy (69.6% vs. 25.0%), whereas in Group 2 the number of patients discharged to a nursing home increased. Conclusions: Falling by elderly patients is correlated to geriatric comorbidities, consequently there was no change in the case numbers in this age group. Strategic measures to avoid COVID-19 infection in hospital setting could include reducing the length of hospital stays by transferring elderly patients to a nursing home as soon as possible and discharging independent, mobile patients to return home.


MedAlliance ◽  
2021 ◽  
Vol 9 (2) ◽  
pp. 8-14

SummaryIntroduction. With the ongoing COVID-19 pandemic, finding new treatments is an extremely important issue. The effectiveness of heliox was previously demonstrated in the complex treatment of patients with various bron-chopulmonary pathologies. Therefore, this method has been recommended for the treatment of pneumonia associated with COVID-19. Purpose. To study the safety and efficacy of inhaled heliox therapy in the treatment of pneumonia in COVID-19. Materials and methods. A sing-le-center prospective study was carried out for the period from 01.12.2020 to 15.02.2021. The study included 91 pa-tients. The patients were divided into two groups: group 1 (using heliox) included 46 people, and group 2 (con-trol) — 45. Inhalations of a heated oxygen-helium mixture heliox (70% helium, 30% oxygen) were carried out using “Ingalit-B2-01” inhaler. Objective (saturation, O2 flow) and laboratory parameters (lactate dehydrogenase, C-reactive protein), as well as chest organs CT data were studied. Differences between groups were determined using the χ2 test, as well as the Mann–Whitney U-test. The p value <0.05 was considered significant. Results. In group 1, side effects developed in 5 (11.3%) patients. These patients refused to further participate in the study. Final number of patients in group 1 — 41. Among patients of group 1, there was a tendency towards a more rapid normalization of lactate dehydrogenase and C-reactive protein, as well as a decrease in oxygen dependence. In group 1, according to CT data, no progression of pneumonia was recorded. In group 2, progression was observed in 6 (13.3%) patients. The overall effectiveness of treatment among patients in group 1 was 100%, among patients in group 2 — 86.7%. The differences between the groups are statistically sig-nificant (p=0.02). Conclusion. The use of inhalations with a heated oxygen-helium mixture heliox (30% oxygen, 70% helium) has shown its effectiveness and safety in the treatment of viral pneumonia (CT1- 2) associated with COVID-19.


2020 ◽  
Author(s):  
Fikret Salık ◽  
Mustafa Bıçak ◽  
Hakan Akelma

Abstract Background Although regional techniques have been suggested more in order to provide postoperative analgesia in inguinal hernia repairs, the ideal method is still controversial. The aim of this study was to evaluate the effect of preemptive transversus abdominis plane (TAP) block, local anesthetic infiltration (LAI) and intravenous dexketoprofen (IVD) on postoperative pain in inguinal hernia repairs. Methods This prospective study included 120 patients with American Society of Anesthesiologists status 1-3 between 18-75 years of age who undergoing elective inguinal hernia surgery under spinal anesthesia. The patients were allocated into three groups: USG-guided TAP block (Group 1, n = 40), USG-guided LAI (Group 2, n = 40) and IVD (Group 3, n = 40) before surgical incision. The mean of tramadol consumption, number of patients in requiring rescue analgesics, duration of postoperative analgesia and complications were recorded for 24 hours postoperatively. VAS scores were evaluated at the 30 minute, 1, 2, 4, 8, 12, 16 and 24th hours. Results There was no significant difference between the postoperative mean tramadol consumption [Group 1: 22.5 mg; Group 2: 20 mg; Group 3: 27.5 (p 0.833)]. The number of patients requiring rescue analgesics was statistically similar (p 0.787). VAS scores at 30. min, 1, 2, 4, 8 and 12 hours were statistically lower in Group 1 than in the other groups (p 0.003, 007, 0013, 0049, 0015, 0021). VAS scores at 30. min and 2. h were lower in Group 1 than Group 2 (p 0.049, 0.037). İn addition, VAS scores at 30 min, 1, 2, 8 and 12 hours were lower in Group 1 than in Group 3 (p 0.003, 0.006, 0.021, 0.017, 0.016). VAS scores at all hours were statistically similar between Group 2 and Group 3. Conclusion This study demonsrated that preemptive transversus abdominis plane block, local anesthetic infiltration or iv dexketoprofen for postoperative analgesia in inguinal hernia repairs had similar effects on postoperative tramadol consumption and number of requiring rescue analgesic patients. However, TAP block provides lower VAS values and excellent analgesia in more patients in the first 24 hours postoperatively.


2021 ◽  
Vol 42 (Supplement_1) ◽  
Author(s):  
S Mohanty ◽  
C Trivedi ◽  
D G Della Rocca ◽  
C Gianni ◽  
B MacDonald ◽  
...  

Abstract Introduction We investigated the ablation success of scar homogenization with combined (epicardial + endocardial) versus endocardial-only approach for ventricular tachycardia (VT) in patients with ischemic cardiomyopathy (ICM) at 5 years of follow-up. Method Consecutive ICM patients undergoing VT ablation at our center were classified into group 1: endocardial scar homogenization and group 2: endocardial +epicardial scar homogenization. Patients with previous open heart surgery were excluded. All patients underwent bipolar substrate mapping with standard scar settings defined as normal tissue &gt;1.5 mV and severe scar &lt;0.5 mV. Non-inducibility of monomorphic VT was the procedural endpoint in both groups. Patients were followed up twice a year for 5 years with implantable device interrogations. Results A total of 361 (Group 1: 291 and group 2: 70) patients were included in the study (mean age: 67 years, male: 88.4%). At 5 years, significantly higher number of patients from group 2 remained arrhythmia-free (figure 1). Of those patients, 87 (45%) and 51 (89%) from group 1 and 2 respectively were off-anti-arrhythmic drugs (AAD) (p&lt;0.001). After adjusting for age, gender, hypertension, diabetes, and obstructive sleep apnea, scar homogenization using endo-epicardial approach was associated with 51% less recurrence compared to the endocardial ablation strategy (Hazard Ratio: 0.49, 95% CI: 0.27–0.89, p: 0.02). Conclusion In this series of patients with ischemic cardiomyopathy and VT, endo-epicardial scar homogenization was associated with a lower need for AAD and a significantly lower recurrence rate at 5-years of follow-up compared to the endocardial ablation alone. FUNDunding Acknowledgement Type of funding sources: None. Figure 1


2019 ◽  
Vol 23 (2) ◽  
pp. 277-282
Author(s):  
V. D. Nemtsova ◽  
I. A. Ilchenko ◽  
V. V. Zlatkina

Due to the growing number of patients with age-related diseases, the aim of the study was to investigate in the changes of aging rate (AR) in patients with comorbid course of arterial hypertension (H), type 2 diabetes mellitus (T2DM) and subclinical hypothyroidism (SH) and to study the features of these changes depending on hormonal imbalance. 118 patients (63 women and 55 men, average age — 53.6±4.3 years) were divided into 3 groups: group 1 (n=37) with isolated H; group 2 (n=42) — with H and T2DM; group 3 (n=39) — with H, T2DM and SH. The investigation program included: measurement of anthropometric parameters (blood pressure, height, body weight (BW), body mass index (BMI)), carbohydrate and thyroid metabolism using standard methods, biological age (BA) by V.P. Voitenko et al. Statistical processing was performed using the Statistica for Windows 8.0 software package. When evaluating AR, physiological aging was found in 8 patients (21.6%) of group 1, in 4 (9.5%) patients of group 2 and 3 (7.7%) of patients in group 3. In the overwhelming majority of the examined patients, premature aging (PA) was noted, however, the acceleration of PA between patients of groups 2 and 3 was not differ significantly (p>0.05). The increase in AR in group 2 patients was accompanied by an increase in BA by 7.2 years, in 3 group patients — by 7.3 years compared with their chronological age. A correlation analysis revealed a positive relationship between BMI and coefficient of aging rate (CAR) (r=0.679; p<0.05); BMI and BA (r=0.562; p<0.05) and CAR and the TSH level (r=0.050; p=0.388) in the 3rd group. Thus, the presence of hypertension and comorbid endocrinopathies — T2DM and SH significantly increases the AR and when assessing the effect of endocrine disorders, the presence of T2DM is more important than SH.


2021 ◽  
Vol 8 (11) ◽  
pp. 3302
Author(s):  
Mahendra P. Singh

Background: It is usual to face clinical complexity in piles patients. They usually present with variable number of pile masses and in different grades of progression. This led to the idea of offering concomitant single stage management in our hemorrhoid patients matching to our criteria. We conducted hospital based descriptive study among the patients coming to my clinical practice falling in grade-1-3.Methods: All the patients falling in grade-1-3 and matching to our criteria were included. The study was conducted   from January 2012 to December 2020. Place of study was basically at two centres: Karamdeep medical centre, Kanpur and Mayo health care, Mohali. Total number of patients were 581. Patients having grade-4 piles and with local co-morbidities were excluded from the study. Modalities of treatment used were sclerotherapy, band ligation and hemorrhoidectomy.Results: Patients managed were divided into four groups – group 1 included patients with piles in grade-1; group-2 included patients having piles in grade-1 and 2; group-3 included patients having piles in grade-1 and 3; and group-4 included patients having piles in grade-1, 2 and 3. Total 952 pile masses were treated in 581 patients. Sclerotherapy was required in 732 (77%) masses, banding in 99 (10.3%) masses and surgery in 122 (12.7%) masses.Conclusions: Concomitant treatment policy proved to be comprehensive way to tackle pile patients of grade-1-3. Mixed and matched method using surgical and non-surgical modalities in a single sitting proved to be beneficial. 86.4% cases were cured this way. Cost of the treatment was economical with lesser complications including local mutilation and better quality of life.


1993 ◽  
Vol 8 (3) ◽  
pp. 131-137 ◽  
Author(s):  
S Dollfus ◽  
M Petit ◽  
JF Menard ◽  
P Lesieur

SummaryThe stability of symptomatology and of eleven schizophrenic diagnostic systems was studied in patients initially in an acute phase (group 1) compared to patients in a residual phase (group 2). The patients were evaluated over a period of 1 year with a standardized checklist and PANSS in both groups. The number of patients included by the Carpenter, Calego, 1CD9, New-Haven, Schneider and Vienne systems decreased significantly between admission and discharge in group 1, whereas no variation was observed in group 1 between discharge and one year later, or in group 2 over a period of 1 year. The instability of schizophrenic diagnostic systems such as New-Haven, Schneider and Vienne could be due to the variation of positive or general symptomatology. The number of patients included by Feighner or Langfeldt did not vary significantly between admission and discbarge in group 1, in spite of a significant decrease in symptomatology, probably because certain criteria, such as duration of illness, hindered the systems from changing. The results showed the importance of specifying in every study on schizophrenia the time of inclusion (admission, discharge) or the phase of illness (acute or residual phase).


2000 ◽  
Vol 28 (1) ◽  
pp. 27-30 ◽  
Author(s):  
R. Madan ◽  
T. Perumal ◽  
K. Subramaniam ◽  
D. Shende ◽  
S. Sadashivam ◽  
...  

This prospective, randomized, double-blinded study evaluated the effect of the timing of ondansetron administration on its antiemetic efficacy in children undergoing elective strabismus surgery. One hundred and twenty children aged one to 15 years, ASA physical status 1 or 2, were randomly allocated to receive intravenous ondansetron 100 μg/kg either at induction (Group 1) or at the end of the surgery (Group 2). All patients had general anaesthesia induced and maintained with nitrous oxide and halothane, muscle relaxation with vecuronium, endotracheal intubation, reversal with neostigmine and glycopyrrolate, and pethidine 0.5 mg/kg analgesia. Episodes of nausea and vomiting were evaluated at 0 to 2, 2 to 6 and 6 to 24 hour intervals by a blinded observer. Demographic data, duration of anaesthesia, type of surgery, incidence of previous postoperative nausea or vomiting and motion sickness and number of patients who developed oculocardiac reflex requiring atropine treatment were similar in both groups. The incidence of emesis in the first 24 hours following surgery was similar in both groups (35% Group 1, 33.3% Group 2, P=1.00). Severity of emesis (median number of emetic episodes, rescue antiemetic requirement and mean time to the onset of first episode of emesis) and mean time to discharge from the post anaesthesia care unit were also similar in the two groups. We conclude that the timing of ondansetron administration either before or after the surgical manipulation of extraocular muscles had similar antiemetic efficacy following strabismus surgery in children.


Sign in / Sign up

Export Citation Format

Share Document