Predicting breast cancer biopsy outcomes from BI-RADS findings using random forests with chi-square and MI features

Author(s):  
Sheldon Williamson ◽  
K. Vijayakumar ◽  
Vinod J. Kadam
Author(s):  
Rini Mayasari Rini Mayasari

ABSTRACT Breast cancer on the mark with the growth or uncontrolled growth of cells that are excessive. Factors affecting the incidence of breast cancer include maternal age, marital status, family history, hormone use, radiation, manarche, and obesity. The purpose of this study was to determine the relationship between Age and Marital Status of Mother with Breast Cancer incidence in the Installation Surgeon General Hospital Dr. Mohammad Hoesin Palembang in 2012. The study design was analytical descriptive with the approach in which the independent variables Cross Sectional Age Mother and Status Perkawianan and dependent variables in breast cancer incidence collected at the same time. The study population was all women who develop breast cancer in the Installation General Hospital Surgery Center Dr. Mohammad Hoesin Palembang in 2012 and samples taken by systematic random sampling with a sample of 130 respondents. The results of univariate analysis of data showed that respondents had breast cancer with a percentage of total 58 (44.61%) while respondents who had not had breast cancer totaled 72 by the percentage (63.39%). Respondents by Age Mothers who are at risk amounted to 34 by the percentage (26.15%) while respondents with Age Mothers who are not at risk amounted to 96 by the percentage (73.85%). Respondents by Marital Status is married and the mother who totaled 69 by the percentage (53.07%) and Marital Status with the status of unmarried women totaled 61 by the percentage (46.03%). Of Chi-Square test results found no significant relationship between Age Mothers with Breast Cancer event in which the p value = 0.007, no significant relationship between Marital Status in Breast Cancer event in which the p value = 0.030. Based on the results of the study, researchers suggest counseling efforts, medical knowledge and improve the quality of health services in order to reduce the incidence of breast cancer..   ABSTRAK Kanker payudara di tandai dengan pertumbuhan atau perkembangan tidak terkontrol dari sel-sel yang berlebihan. Faktor-faktor yang mempengaruhi kejadian kanker payudara antara lain umur ibu, status perkawinan, riwayat keluarga, penggunaan hormon, radiasi, manarche, dan obesitas. Tujuan penelitian ini adalah untuk mengetahui hubungan antara Umur Ibu dan Status Perkawinan dengan kejadian Kanker Payudara di Instalasi Bedah Rumah Sakit Umum Pusat Dr. Mohammad Hoesin Palembang Tahun 2012. Desain Penelitian ini adalah deskriptif analitik dengan pendekatan Cross Sectional dimana variabel independen Umur Ibu dan Status Perkawianan dan variabel dependen kejadian kanker payudara dikumpulkan dalam waktu yang bersamaan. Populasi penelitian ini adalah semua ibu yang menderita kanker payudara di Instalasi Bedah Rumah Sakit Umum Pusat Dr. Mohammad Hoesin Palembang Tahun 2012 dan sampel di ambil secara systematic random sampling dengan jumlah sampel 130 responden. Hasil analisis univariat data menunjukan responden yang mengalami Kanker Payudara berjumlah 58 dengan persentase sebesar (44,61%) sedangkan responden yang tidak mengalami Kanker Payudara berjumlah 72 dengan persentase (63,39%). Responden dengan Umur Ibu yang beresiko berjumlah 34 dengan persentase (26,15%) sedangkan responden dengan Umur Ibu yang tidak beresiko berjumlah 96 dengan persentase (73,85%). Responden dengan Status Perkawinan ibu yang berstatus kawin berjumlah 69 dengan persentase (53,07%) dan Status Perkawinan ibu yang berstatus tidak kawin berjumlah 61 dengan persentase (46,03%). Dari hasil uji Chi-Square didapatkan ada hubungan bermakna antara Umur Ibu dengan kejadian Kanker Payudara dimana p Value = 0,007, ada hubungan bermakna antara Status Perkawinan dengan kejadian Kanker Payudara dimana p Value = 0,030. Berdasarkan hasil penelitian, peneliti menyarankan upaya-upaya penyuluhan, pengetahuan tenaga medis dan meningkatkan mutu pelayanan kesehatan guna menurunkan angka kejadian kanker payudara.


2018 ◽  
Vol 3 (3) ◽  
pp. 575
Author(s):  
Neila Sulung ◽  
Rizki Yananda ◽  
Adriani Adriani

<p>Cancer is one of the leading causes of death worldwide. In Indonesia every year 1: 3 women per 1000 population are affected by breast cancer. Breast cancer is a cancer that attacks most women. The incidence of breast cancer is currently estimated at 39 per 100,000 population in 2008. The purpose of this study was to determine the factors associated with the risk of female breast cancer in surgical outpatient poly patients at Dr. Achmad Mochtar, Bukittinggi City. This study uses descriptive analytic method with a case control approach. The sampling technique in this study was accidental sampling. The sample in this study were all women diagnosed with breast cancer, amounting to 50 cases and 50 controls with data processing through computerization. The instrument used in this study is a questionnaire. Data analysis was performed using Chi-Square test (α = 0.05). The results showed that the factors associated with the incidence of breast cancer were genetic (p = 0.009), menarche (p = 0.014), menopause (p = 0.016), hormonal contraception (p = 0,045), obesity (p = 0,043), and high food fat (p = 0.028).  Conclusions of the study are factors related to the risk of breast cancer incidence are genetic, menarche, menopause, hormonal contraception, obesity and high-fat foods.<br /> </p><p>Penyakit kanker merupakan salah satu penyebab kematian utama di seluruh dunia. Di Indonesia setiap tahun 1:3 wanita per 1000 penduduk terserang kanker payudara. Kanker payudara merupakan kanker yang paling banyak menyerang perempuan. Angka kejadian kanker payudara saat ini diperkirakan 39 per 100.000 penduduk pada tahun 2008. Tujuan penelitian ini adalah untuk mengetahui faktor-faktor yang berhubungan dengan risiko kanker payudara wanita pada pasien poli rawat jalan bedah di RSUD Dr. Achmad Mochtar Kota Bukittinggi. Penelitian ini menggunakan metode <em>deskriptif analitik</em> dengan pendekatan <em>case control</em>. Teknik pengambilan sampel dalam penelitian ini adalah <em>accidental sampling.</em> Sampel dalam penelitian ini adalah semua wanita yang terdiagnosis kanker payudara, berjumlah 50 kasus dan 50 kontrol dengan pengolahan data melalui komputerisasi. Instrument yang digunakan dalam penelitian ini adalah lembar kuisioner. Analisis data dilakukan menggunakan uji <em>Chi-Square </em>(α=0,05). Hasil penelitian menunjukkan faktor yang berhubungan dengan kejadian kanker payudara adalah genetik (p=0,009), <em>menarche</em> (p=0,014;), <em>menopause</em> (p=0,016), kontrasepsi hormonal (p=0,045), <em>obesitas </em>(p=0,043), dan makanan tinggi lemak (p=0,028). Simpulan penelitian adalah faktor yang berhubungan dengan risiko kejadian kanker payudara adalah genetik, <em>menarche, menopause,</em> kontrasepsi hormonal, <em>obesitas</em> dan makanan tinggi lemak.</p>


2021 ◽  
Vol 20 ◽  
pp. 153473542098834
Author(s):  
Abdolazim Sedighi Pashaki ◽  
Kamal Mohammadian ◽  
Saeid Afshar ◽  
Mohammad Hadi Gholami ◽  
Abbas Moradi ◽  
...  

Objective: Fatigue associated with malignant conditions and their treatments is a disabling condition. This trial assessed the anti-fatigue effects of melatonin coadministration during adjuvant treatment of patients with the breast cancer. Material and Methods: Patients with breast cancer were randomly assigned to receive melatonin or placebo during adjuvant chemotherapy and radiotherapy. Thirty-seven patients were randomly enrolled in each group. The mean ages of patients in the intervention and control groups were 50.47 ± 10.79 and 46.05 ± 10.55 years, respectively ( P = .223). The intervention group received oral melatonin (18 mg/day) from 1 week before until 1 month after the adjuvant radiotherapy. The level of fatigue was assessed before and after intervention using Brief Fatigue Inventory (BFI) in both groups. To analyze data, the Student’s t-test and the Chi-square test were used at a significance level of P ≤ .05. Results: The BFI score was similar before the intervention in both groups, however, after the intervention, it was significantly lower in the melatonin group ( P < .001). Moreover, the frequency of severe fatigue in the melatonin group was significantly lower than in the placebo group after intervention (42.1% vs 83.3%, P < .001). Conclusion: Coadministration of melatonin during adjuvant chemotherapy and radiotherapy of women with breast cancer decreased the levels of fatigue associated with the malignant condition and its treatments.


2021 ◽  
Author(s):  
Xuan Thi Nhu Ha ◽  
Sureeporn Thanasilp

Background: The patient’s quality of life immediately after mastectomy usually receives less attention than the quality of life after three months, six months, or a year. It is because the focus is mainly on surgical complications. Many instruments measure the quality of life from three months onwards. Still, the quality-of-life instruments right after postmastectomy is not yet verified.Objective: This paper aimed to test the reliability and validity of the Quality-of-Life Index Vietnamese version (QOLI-V) in Vietnamese women with breast cancer three weeks postmastectomy.Methods: The descriptive cross-sectional study was designed to analyze the psychometric properties of a Vietnamese version of the modified Quality of Life Index. The modified process was conducted after granting permission from the original authors. The content validity of the modified index was examined by five experts. Brislin’s model was used for the translation process. The 26-item QOLI-V was tested in 265 patients with breast cancer stage II three weeks postmastectomy who expected to have a poorer quality of life score. The reliability of the index was measured by Cronbach's alpha. The construct validity was examined using confirmatory factor analysis (CFA).Result: The content validity index results showed that the lowest I-CVI was .80 and the highest was 1.00. S-CVI/Ave was 0.95, and S-CVI/UA was 0.76. The Cronbach's alpha of QOLI-V was .84, which was considered acceptable. Most of the 26 items featured the correct item-total correlation of .30 to .60. There were only two items correlated with the total scale at .18, and the item with the lowest correlation (.06) was deleted from the item set.  The CFA of model 1 with 26 items was not an ideal fit with the data, with Chi-Square/df = 2.15, CFI =.815, GFI = .853, TLI=.792 RMSEA =.066. After deleted an item #general quality of life, and the CFA of model 2 was conducted on the 25-item index. The final result indicated the improvement of the model fit, with Chi-Square/df =2.26, CFI=.852, GFI=.814, TLI=.790, RMSEA=.069.Conclusion: The 25-item QOLI-V version is considered valid and reliable to measure the quality of life of Vietnamese women with breast cancer three weeks postmastectomy. Nurses and midwives could use this instrument to measure the quality of life of the patients, and the patients could use it for self-assessment.Funding: This study was funded by the 90th Anniversary of Chulalongkorn University Scholarship, Chulalongkorn University, Thailand.


2021 ◽  
Vol 39 (15_suppl) ◽  
pp. 12053-12053
Author(s):  
Marisa C. Weiss ◽  
Stephanie Kjelstrom ◽  
Meghan Buckley ◽  
Adam Leitenberger ◽  
Melissa Jenkins ◽  
...  

12053 Background: A current cancer diagnosis is a risk factor for serious COVID-19 complications (CDC). In addition, the pandemic has caused major disruptions in medical care and support networks, resulting in treatment delays, limited access to doctors, worsening health disparities, social isolation; and driving higher utilization of telemedicine and online resources. Breastcancer.org has experienced a sustained surge of new and repeat users seeking urgent information and support. To better understand these unmet needs, we conducted a survey of the Breastcancer.org Community. Methods: Members of the Breastcancer.org Community were invited to complete a survey on the effects of the COVID-19 pandemic on their breast cancer care, including questions on demographics, comorbidities (including lung, heart, liver and kidney disease, asthma, diabetes, obesity, and other chronic health conditions); care delays, anxiety due to COVID-related care delays, use of telemedicine, and satisfaction with care during COVID. The survey was conducted between 4/27/2020-6/1/2020 using Survey Monkey. Results were tabulated and compared by chi square test. A p-value of 0.05 is considered significant. Data were analyzed using Stata 16.0 (Stata Corp., Inc, College Station, TX). Results: Our analysis included 568 breast cancer patients of whom 44% had ≥1 other comorbidities associated with serious COVID-19 complications (per CDC) and 37% had moderate to extreme anxiety about contracting COVID. This anxiety increased with the number of comorbidities (p=0.021), age (p=0.040), and with a current breast cancer diagnosis (p=0.011) (see table). Anxiety was significantly higher in those currently diagnosed, ≥65, or with ≥3 other comorbidities, compared to those diagnosed in the past, age <44, or without other comorbidities. Conclusions: Our survey reveals that COVID-related anxiety is prevalent at any age regardless of overall health status, but it increased with the number of other comorbidities, older age, and a current breast cancer diagnosis. Thus, reported anxiety is proportional to the risk of developing serious complications from COVID. Current breast cancer patients of all ages—especially with other comorbidities—require emotional support, safe access to their providers, and prioritization for vaccination.[Table: see text]


2021 ◽  
Vol 39 (15_suppl) ◽  
pp. 12055-12055
Author(s):  
Christine M. Duffy ◽  
Harish Saiganesh ◽  
Stavroula Chrysanthopoulou ◽  
Camille Higel-Mcgovern ◽  
Don S. Dizon

12055 Background: While the Commission on Cancer has eliminated strict quotas for accreditation, Survivorship Care Plans (SCPs) and/or Survivorship Care Visits (SCV) at treatment completion are encouraged. However, who receives a SCP or SCV, whether it impacts care, and impact of distress on care is unknown. We examined the provision of survivorship care at the Lifespan Cancer Institute (LCI) to determine (1) clinical and distress thermometer scores (DTS) association with SCPs and SCVs; (2) impact of SCV visits on specialty referrals, and (3) demographic and clinical predictors of receipt of SCP and SCV. Methods: We retrospectively reviewed EMR records on 1,960 patients at LCI between 2014-2017 for SCPs and SCVs and extracted demographics, treatment variables, and distress scores. We used T-test or Wilcoxon rank test and Chi-square tests for evaluating the bivariate associations of SCP and SCV with continuous and categorical factors respectively. We fit logistic regression models to assess the adjusted effect of these factors on receipt of SCP and SCV independently. All analyses were performed in R v4.0.2. Results: The mean age was 63.9 (SD=11.8), 67% were female, 51.2% were married or partnered. Breast (38.8%), lung (17.6%), and prostate (13.7%) were the most common cancers. DTS were recorded in 64% with mean of 3.88(SD=3.05): distress was higher in women (4.36, SD=3.01), breast cancer pts (4.53, SD=3.07), gyn (4.22, SD=3.07), pancreatic (4.12, SD=3.41) and anal cancers (4.52, SD=3.47) and in those with Stage IV disease (5.33, SD=3.43). SCPs were completed in 740 (37.8%) patients and of those 65.9% had a SCV. SCV were associated with more specialty referrals for psychiatry, physical therapy, nutrition, and sexual health but not smoking cessation or fiscal services. DTS were associated with increased referral to psychiatry only. The adjusted models (table) showed odds of receiving a SCP were higher in those younger, and having breast cancer v all other cancers, with prostate having lowest odds. For receipt of SCV, odds were higher in those younger, female, and having breast cancer, with prostate and lung having the lowest odds. Conclusions: Gender, age and type of cancer are significant predictors of receipt of SCP and SCV. SCP and SCV patterns may represent patient preferences, but practice patterns and unconscious biases may also play a part suggesting areas for further research and outreach.[Table: see text]


2021 ◽  
Vol 39 (15_suppl) ◽  
pp. e13065-e13065
Author(s):  
Qian Dong ◽  
Mi Zhang ◽  
Da Jiang

e13065 Background: To analyze the correlation between tumor size and metastatic site in first-diagnosed stage IV breast cancer patients. Methods: Stage IV breast cancer patients diagnosed from 2010 to 2015 were screened by the Surveillance, Epidemiology, and End Results (SEER) database. The characteristics of clinical variables were represented by a frequency table, and the Chi-square test was used for comparison. At the same time, the Chi-square test was used to analyze the relationship between tumor size and organ metastasis. Correlation between tumor size and the prognosis of patients was contributed by KM curve and Log-rank test. Results: Regardless of tumor size, the proportion of bone metastasis was higher and brain metastasis was lower in breast cancer patients. There were significant differences in the site of metastases based on different subtype. Luminal A and Luminal B breast cancer had the highest proportion of bone metastases; brain metastasis accounted for the highest proportion in triple-negative breast cancer (TNBC); while the incidence of liver metastasis was the highest in Her-2(+) breast cancer. At the same time, the results indicated that Luminal A breast cancer with a tumor size > 5 cm was more likely to develop multi-site metastasis and lung metastasis, while Luminal B breast cancer with a tumor size ≤ 5 cm was more likely to develop liver metastasis. The results also revealed that TNBC patients with a tumor size of 0 - 2cm were more likely to develop bone metastasis than those with a tumor size > 5 cm, and the incidence of lung metastasis in triple-negative patients showed an increasing trend with the increase of tumor size. Conclusions: Based on subtype, we found that there was a significant difference between tumor size and metastatic site in patients with stage IV breast cancer, and the difference was statistically significant. This study provided evidence-based basis for decision-making of stage IV breast cancer treatment.


2012 ◽  
Vol 78 (6) ◽  
pp. 669-674 ◽  
Author(s):  
Patrick D. Glasgow ◽  
Nikhil Satchidanand ◽  
Gopal Chandru Kowdley

The rate of micrometastatic disease (MMD) to nonsentinel lymph nodes (NSLNs) has been shown to vary considerably in the literature. We identified patients with breast cancer with MMD (N1mi) and measured the incidence of NSLN involvement. We then compared these patients with those who had no metastasis to the SLN (N0) and those who had macrometastasis to the SLN (N2) in an attempt to better understand the behavior of patients with N1mi positivity. A retrospective analysis was conducted on 574 patients with invasive breast cancer between January 2000 and December 2007. Patients were stratified into three groups: no metastasis (N0), MMD (N1mi), and macrometastasis (N2). Chi square analysis and logistic regression models using SPSS software were applied to determine significance between groups. MMD rate was 7.7 per cent (44 of 574). Of this subset of patients, 33 underwent completion axillary dissection, and only two were found to have NSLN-positive disease. Statistical significance was achieved for NSLN positivity when comparing all three nodal groups against one another (χ22,572 = 337.084, P = 0.000). Logistic regression showed multifocality and lymphovascular invasion to be significant predictors of NSLN metastasis. NSLN positivity in patients with MMD acts similarly to node-positive disease and therefore cannot completely exclude axillary dissection from therapeutic algorithm.


1998 ◽  
Vol 13 (3) ◽  
pp. 139-144 ◽  
Author(s):  
T.M. Maguire ◽  
S.G Shering ◽  
C.M. Duggan ◽  
E.W. McDermott ◽  
N.J. O'Higgins ◽  
...  

Cathepsin B (CB) is a thiol-stimulated protease implicated in cancer invasion and metastasis. Other proteases involved in cancer spread such as urokinase-type plasminogen activator (uPA) and cathepsin D have previously been shown to be prognostic markers in breast cancer. CB was assayed by ELISA in 193 patients with primary breast cancer. CB levels were significantly higher in both primary and metastatic breast tumors than in fibroadenomas (p=0.0001). In the primary carcinomas, CB levels showed no significant correlation with either nodal status, tumor size or estrogen receptor (ER) status. Patients with primary breast cancers containing high levels of CB had a significantly shorter disease-free interval (p=0.01, chi-square=6.61) and overall survival (p=0.014, chi-square=6.08) than patients with low levels of the protease. However, in multivariate analysis, using nodal status, tumor size, ER status and urokinase plasminogen activator (uPA), CB was not an independent prognostic marker. In contrast, nodal status, ER status and uPA were prognostic in multivariate analysis. In conclusion, CB, like certain other proteases implicated in cancer metastasis, correlates with poor outcome in patients with breast cancer. These results thus support the evidence from model systems linking CB to cancer spread. Inhibition of CB expression or activity might therefore be exploited for anti-metastatic therapies.


Sign in / Sign up

Export Citation Format

Share Document