 Hello, everybody. My name is Dr. Anumitra Chakraborty. I am currently a third-year resident in the Department of Radiology and Imaging, Northeastern Indira Gandhi Regional Institute of Health and Medical Sciences, Shillong, Mighalaya. My paper is titled, Diffusion Weighted Imaging and Magnetic Resonance Spectroscopy, Predicting Malignancy in Indeterminate Ovarian Tumors. It is done under the guidance of Dr. C. Daniela, Professor and HOD, Department of Radiology and Imaging, Dr. Pranjil Fukan, Additional Professor, Department of Radiology and Imaging, Dr. Donbuck Lang-Linzer, Associate Professor, Department of Radiology and Imaging, Dr. Keeley-Bharis, Associate Professor, Department of Surgical Oncology, Dr. Rituparna Das, Assistant Professor, Department of Obstetrics and Gynaecology, and Dr. Jaya Mishra, Associate Professor, Department of Pythology. So, coming to the background and overview of the disease and the imaging at hand, so, out of all the gynecological malignancies, ovarian cancer is the third most common, the most common cancer being cervical carcinoma followed by endometrial carcinoma. However, out of all these gynecological malignancies, ovarian cancer carries the worst prognosis and the highest mortality rate with very low survival rates. The C.R. cancer statistics fact sheets say that the relative 5-year survival rate is 48% with the median age of diagnosis approximately 63 years. This poor prognosis is attributable mainly to the asymptomatic course of the disease with delayed onset of symptoms and a lack of proper screening modalities. Imaging plays a very vital crucial role in the management of these patients because it aids in the detection, the differentiation of benign borderline and invasive malignant ovarian masses, aids in the staging and in the prognostication of the same. So, coming to the various modalities at hand, we have CT, which has a short acquisition time and can take whole body scans very rapidly with minimal artifacts. It has chosen to detect as well as stage the diseases but has a limited role in characterization and carries the risk of radiation. Transvaginal ultrasound and contrastal MRI have high sensitivity of 97% and 100% respectively in identifying the solid components within the masses. An MRI offers a overall higher accuracy of 93%. In MRI, diffusion-coded imaging gives us a non-invasive evaluation of the extent of microscopic diffusion within the biologic tissues and can differentiate between the benign epithelial tumors with solid components from the malignant with high sensitivity and specificity. Magnetic resonance spectroscopy gives us information about the anatomy as well as helps to provide a quantification of the metabolic features of the tumor. Combining the MRI, conventional MRI, with TWI and MRS findings to distinguish between the benign and malignant while attempting to give a diagnosis of the histopathological type of the ovarian carcinoma may offer a faster non-invasive means as compared to a tissue diagnosis. So the aims and objectives of this study was mainly to assess the diagnostic accuracy of diffusion-coded imaging, DWI and magnetic resonance spectroscopy in differentiating between benign and malignant ovarian masses. So coming to the methodology, our study type was a prospective observational study. It was approved by the Institutional Ethics Committee of Negrims. The sampling population was actually prospective cases of indeterminate ovarian masses who were screened on ultrasound and then underwent an MRI pelvis with diffusion-coded imaging, MRS and contrast enhancement in the Department of Radiology and Imaging at Negrims and subsequently the ones who went on to obtain a tissue diagnosis were taken up for the statistical analysis. So finally a sample size of 40 consenting adult females were ultimately screened on ultrasound and subsequently went underwent the MRI and the histopathological confirmation. The gold standard of our study was the histopathology. So the inclusion criteria included adnexal or ovarian masses who had the following features. Purely cystic masses more than 4 cm in size or solid masses, complex cystic masses or complex solid masses with small or large solid components. We excluded small functional cysts which were less than 4 cm, simple cyst less than 4 cm, endometrium mass which were confirmed by any of the diagnostic modalities that included ultrasound, MRI or laparoscopy. Adnexal masses who had already done a pre-existing HP diagnosis like a biopsy were excluded from our study and general contraindications to MR including implants and metallic artifacts etc. were already excluded from the study. So the patients who had a clinical suspicion of adnexal or ovarian neoplasms were subsequently screened on the ultrasound. Out of them 40 patients were selected with indeterminate lesions satisfying the inclusion criteria. They underwent a contrast in their MRI with DWI and MRS using the Siemens 1.5 Tesla Magnetum Avento MR scanner in an institute. The mean ADC values of the solid and cystic components were measured. For the MRS the spectral reconstruction was done with the aid of the software at the workstation of the MR device. Additionally all the conventional MRI sequences were also obtained. The images were all analyzed in the presence of a senior radiologist and for the gold standard the patients underwent biopsy or surgical interventions followed by HP diagnosis and finally the statistical analysis was done on the SPSS version 22 software. So coming to a few of our cases the first case was a 52 year old female who was diagnosed as a case of serious cystidinoma after imaging and HPE. So the imaging showed this is the T2 weighted axial image, the T1 weighted axial image, the T1 weighted post contrast fat suppressed axial image. We can see the multi-loculated cystic pelvic right adnexilation with some thick septation having intermediate signal on the T2 weighted imaging. The diffusion weighted image, the trace diffusion image showed no evidence of a high signal in that corresponding solid component and the ADC map did not reveal any reduced values or restricted diffusion. The multi-voxel magnetic resonance spectroscopy that was done with a TE value of 135 showed an NAA peak and a lipid peak with a lactate inversion. However there was no elevation of the choline or the creatinine and the choline creatinine ratio was low. The HP images that we have included here include the 40X and the 200X hematuxiline eosin stain slides. And we can see the papillary projections, multiple papillary projections and we can see a single layer of ciliated columnar epithelial cells lining the same. So our second case was a 45 year old female who was diagnosed as a serious cystidinocarcinoma. So this is the T2 weighted axial image, the T1 weighted axial image, the T1 weighted post contrast fat suppressed axial image and all three show a complex abdominal pelvic lesion with enhancing solid portion and cystic spaces and acitis. This is the trace diffusion image showing intermediate signal and the ADC map showing low values on ADC meaning restricted diffusion in the solid component. The multi-voxel magnetic resonance spectroscopy done with a TE value of 135 showed a choline peak, no creatinine peak. There was an NAA peak, lactate inversion and a lipid peak. The choline creatinine ratio was high. The 40X HND and the 200X HND slide images showed multiple papillary projections with uniform population of small cells with scan cytoplasm and mild nuclear ATP. Our third case was a 55 year old female who was diagnosed as a case of miscellaneous cystidinoma based on the imaging and the HP. So this is the T2 weighted fat suppressed societal image, the T1 weighted fat suppressed axial image and the T1 weighted post contrast fat suppressed societal image showing a complex cystic mass abdominal pelvic lesion with enhancing wall, thick wall and thick septa. The diffusion weighted imaging and the ADC map showed no significant restricted diffusion. The MRS at TE 135 value showed an NAA peak and no other significant metabolic peaks were noted. The 40X and the 200X HND slides showed a cystic neoplasm with multiple cysts and clans and papillary projections lined by a single layer of planned miscellaneous epithelium. Our fourth case was a 27 year old female diagnosed as miscellaneous cystidinoma. This is the T2 weighted fat suppressed axial image. The T2 weighted non fat suppressed axial image, the T2 weighted coronal image showing a complex cystic abdominal pelvic lesion with solid components, a cytase. The diffusion weighted imaging and the ADC maps showed restricted diffusion with low ADC value in the solid portion. The multipoxyl spectroscopy at TE value of 30 showed a choline peak, lipid lactate peak, high choline creatinine ratio. There was no NAA peak in this patient. The 40X HND slide shows multiple tumor cells reaching up to the capsule and the 200X image we can see some glands with musins secreted in between and stromal invasion with papillary atypical moderately pleomorphic stratified cells. So coming to the results, we have summarized them in graphs and charts. So we had a total sample size of 40 and out of them, 27 turned out to be malignant, 13 turned out to be benign. In the malignant group, almost 10 had mixed composition and in the benign group, 10 had cystic composition. These were the normal parameters that we took into consideration, the margins, the wall, septa, vegetations, solid portions, a cytase, deposits, lymphadenopathy, pelvic organ involvement or sidewall involvement and intermediate signal of the solid component on T2. So in the malignant group, we saw that nearly 20 had irregular margins, 22 had thick septa, 16 had thick wall and 23 had intermediate signal. In the benign group also, we noticed that 9 had intermediate signal on the T2 weighted imaging. Other parameters were not so significant. Moving on to the other statistical slides, we can see the ADC values we have taken here, the trace diffusion image. We saw the intermediate signal in nearly 22 of malignant cells and in 4 of the benign cells. And the T test that was done for the equality of means showed a p-value of 0.001, which was significant. In the malignant group, the mean ADC value was 0.809 plus minus 0.054, with a standard division of 0.283, median of 0.805 and a confidence in the 95% confidence level of 0.697 to 0.921. The benign group had a mean ADC value in comparison higher of around 1.511 plus minus 0.153, standard division of 0.555, a median of 1.685 and a confidence interval as here. And these are the box whisker plot that showed the significant difference of the median values of the malignant and the benign groups with a negative skew in the benign group. So regarding the metabolite peaks, choline peak was seen in nearly 23 of the malignant lesions. Lipid peak was seen in 16 malignant lesions and lactate in 18 of the malignant lesions. Creatine peak was seen in only 4 of the malignant lesions but not seen in any of the benign lesions. The choline peak showed the highest sensitivity of 85.2% with a specificity of 92.31% in the malignant group. And the Fisher exact test showed a p-value of less than 0.05. The creatine peak showed highest specificity of 100% and the highest positive predictive value of 100% but had a low sensitivity of 14.81%. In the malignant group, the choline creatine integer ratio, mean ratio was around 2.86 which was significantly higher than the mean ratio of the benign group which was 1.052 with 95% confidence intervals as given here, median of 2.97 in the malignant and 1.010 in the benign groups respectively. So coming to the statistical test that we employed in the spectroscopy findings, the creatine peak showed significant relationship with the malignant group with p-value of less than 0.05. Choline also showed statistical significance. However, NA lipid and lactate did not show much significance. In the test of normality, we found that the choline creatine ratio had a higher significance, statistical significance in the malignant group with relatively no significance in the benign group. And in the source tests in the choline creatine ratio, we found a p-value of overall less than 0.0001 which was highly significant. And if we compared the determination of malignancy by choline creatine ratio and the mean ADC values in the solid component, we saw a strong correlation with a linear relationship in the regression charts. The box whisker plots in the malignant and the benign group also showed significant difference in the median of the choline creatine ratio in both the groups. Comparing the diagnostic accuracies, we found that the sensitivity significantly increased after on addition of the diffusion weighted imaging and the spectroscopy findings to conventional MRI parameters with a peak of 96.3%. And the specificity also went on increasing accordingly. The positive likelihood ratio and the negative likelihood ratios improved with the addition of DWI and MRS and the negative predictive value significantly increased with the addition of both DWI and MRS. However, positive predictive value was still the highest in the conventional MRI parameters. Coming to the discussion, our findings in our study corroborated with the studies done by Wenhua Lee et al. regarding DWI in the malignant masses. So, they said that the strong DWI signal was associated with the solid portion of the tumors while the cystic portion produced a weaker signal. And the lower mean ADC values in the solid component were predominantly associated with malignant ovarian surface epithelial saturated regression mass, which were similar to what we found. A study conducted by Isabel Thamasin Nagara et al. showed that the presence of solid component with intermediate T2 signal and high signal intensity at B value of 1000 was associated with the positive likelihood ratio of 4.5 for a malignant and nexal tumor. The DWI signal intensity was an accurate tool for predicting benignity of the complex nexal masses. Coming to the spectroscopy findings, Isabel Thamasin Nagara et al. said that the identification of a choline peak demonstrated sensitivity of 89% and specificity of 84% regarding the differentiation of malignant from benign ovarian pathologies. Ma et al. said that choline-creating integral ratio was significantly higher in the malignant groups compared to the benign groups. Values were much higher than what we found in our study. Studies done by Stanwell et al. and Forstner et al. showed that choline-creating integral ratio of more than 3 indicated that the tumor was malignant in nature. And the value of less than 1.5 implied that the tumor was benign in nature. And finally, the study done by Mansour et al. showed that the addition of MRS to the traditional MR imaging increased the overall diagnostic accuracy of the MR modality in the characterization of the ovarian neoplasm and screening of malignancies. So in conclusion, we can see that the diffusion-rated MRI appears to offer a reliable method to differentiate malignant and benign ovarian pathologies based on the estimation of cellularity in the solid component. Proton magnetic resonance spectroscopy also provides good biochemical information and offers good diagnostic accuracy when screening the different ovarian masses with choline peak and choline-creating ratio being the most reliable parameters. The addition of DWI and MRS to the conventional MRI sequences in the addition in the evaluation of adnexal masses greatly increases the diagnostic accuracy. So these were my references. Thank you.