{"cells":[{"metadata":{},"cell_type":"markdown","source":"# Understanding Pulmonary Fibrosis Progression    \n\n  Pulmonary fibrosis is a progressive disease that naturally gets worse over time. This worsening is related to the amount of fibrosis (scarring) in the lungs. As this occurs, a person's breathing becomes more difficult, eventually resulting in   shortness of breath, even at rest.\n\n  Patients with pulmonary fibrosis experience disease progression at different rates. Some patients progress slowly and live with PF for many years, while others decline more quickly.\n\n  There is no cure for pulmonary fibrosis, but treatments can slow the progression of the disease in some people. \n  \n","execution_count":null},{"metadata":{"trusted":true},"cell_type":"code","source":"from IPython.display import YouTubeVideo\nYouTubeVideo('QdwuHKwOLRU', width=800, height=300)\n","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## What is Forced Vital Capacity and how does Lung Function change with age, gender and smoking habbits.\n\nForced vital capacity (FVC) is the amount of air that can be forcibly exhaled from your lungs after taking the deepest breath possible, as measured by spirometry.\n\nForced expiratory volume is the most important measurement of lung function. It is used to:\n\n    1) Diagnose obstructive lung diseases such as asthma and chronic obstructive pulmonary disease (COPD).\n    \n    2) See how well medicines used to improve breathing are working.\n  \n    3) Check if lung disease is getting worse. Decreases in the FEV1 value may mean the lung disease is getting worse.\n    \n### Lung function can be divided into three categories: spirometry to assess the dynamic flow rates: forced expiratory volume in one second (FEV1), forced vital capacity (FVC), and FEV-1/FVC ratio.\n\n\n## Change in Lung Function with Age [Non Smokers]\n\n\n![agevlf.jpg](attachment:agevlf.jpg)\n\nAge-related decline in forced expiratory volume in one second (FEV1)% predicted plotted as % of maximal at age 20 years against age.\n\nsource: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2695176\n\nEffect of aging on respiratory system physiology and immunology\nGulshan Sharma1 and James Goodwin2\n\nClin Interv Aging. 2006 Sep; 1(3): 253–260. Published online 2006 Sep. doi: 10.2147/ciia.2006.1.3.253\n\n## Change in Lung Function with Age [Smokers vs Non Smokers]\n\n\n![smvage.gif](attachment:smvage.gif)\n\nThe natural history of lung function decline. Smokers who are susceptible to lung injury experience an increase in the rate of age-related loss in FEV1 compared with nonsmokers (red, green, and blue lines). After lung function declines to threshold levels, clinical symptoms develop (black dotted lines). When a smoker stops smoking, the rate of FEV1 loss again approximates to that of a nonsmoker (blue dotted line). (FEV1 = forced expiratory volume in one second.)\n\nsource: https://www.aafp.org/afp/2006/0215/p669.html\n\n\n## Change in Lung Function with Age [Males Vs Females]\n\n<img align=\"center\" width=\"900\"  src=\"https://github.com/tkrsh/osic-pulmonary-fibrosis-progression/blob/master/files/ssd.png?raw=true\"> \n  \n### The most conspicuous group differences at baseline for women were also mainly related to smoking, and the presence of respiratory symptoms: smokers had a greater risk of being in the ‘accelerating decline’\n\n\n\nsource: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0197250\n\n    ","attachments":{"agevlf.jpg":{"image/jpeg":"/9j/4AAQSkZJRgABAQEASABIAAD/2wBDAAEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQH/wAALCAFQAhYBAREA/8QAHwABAAICAgMBAQAAAAAAAAAAAAgJBwoFBgEDBAIL/8QAPxAAAAYDAAEBBAgFBAEDAwUAAAIDBAUGAQcICRESExiYChQhVlh41tcVNje3uBYXMVFBIiNxGSRhJzJCUpH/2gAIAQEAAD8A3+AAVTeW3yH3Dx+88zFl1DovYW+t72ijbRsuuYSuVWRk9cUqK1XEQsteNlbtuCJ20ZTKJT4+xRrxNB48RlLjKKN61Aky6cLu2XvvXVXY8j45Odd9cyaIgN6dU9BU/n3DStOjPInV9DlNvQMVJWvaN5TayaM4lrPXqbp3JSTCIfrzTlPLBi3OoZZRYsV+cvKZ0DSpXyl6s7sq2mJvYXjK1zR9wT2xOaz2eL1psSpbB1lYL/G1LLK+SEnIV+9xC1cWjZIq7/6qfMmiTDFE8ass+6jwR5Q+1b10tydpbubXXPdWhPIXy3a+oOY3ekF70nO0JSnJxVhltTbRRtzyRZTcqehzzCcbWWDVj2n8RZuo/LVX6ynhtsUAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAhV5IKPbtl8B9k6+oFclrfeLpzbt+sVKrQLNSQmrDYJmlS7GLh4pkjgyrt/IPFkWzZunjJ1VjkIXGc5wK9ehNrdw8beF3QMdyVzNsrb3ZH+xOitLw1Ar9KkLNN6lsLjWLdrab9bqw3KVUyWvf4K7bJRzwybZxb3cDHSGMslnRRCTgPmmw7w8a/fHEkhxd2Zztu7e+nrXKbY6Q7gioqDsXVfQe3a5Zo+Ssac/ETU5IJRdcnIximnCqny1q1cmI1NHL545k3brj/H7pvrroPtvxo3/AHXyTubl2teLPh7YujNnWbbEfBxsFtTeV1gq5rBrC6hXj5aRPcKa3gq6+t6lwYp4jiJ5asFTpqP2Cjva8k4yNm42RhpmOYy8PLsXcZLRMm0bv42TjX7dRo+jpFi7TVavWL1qqq2dtHKSrdy3VURWTOmcxc1tbx445DitnccMIvlbm+NY2XpC0QljZsNHaxZtZ+GQ5D6psaERNt29XTRlYxGwwEDPJMH5F2qczCREoRIr6NZroyQ+B3ir8H/Lny/6n/SQfA7xV+D/AJc+X/U/6SD4HeKvwf8ALny/6n/SQfA7xV+D/lz5f9T/AKSD4HeKvwf8ufL/AKn/AEkHwO8Vfg/5c+X/AFP+kg+B3ir8H/Lny/6n/SQfA7xV+D/lz5f9T/pIPgd4q/B/y58v+p/0kHwO8Vfg/wCXPl/1P+kg+B3ir8H/AC58v+p/0kHwO8Vfg/5c+X/U/wCkg+B3ir8H/Lny/wCp/wBJB8DvFX4P+XPl/wBT/pIPgd4q/B/y58v+p/0kHwO8Vfg/5c+X/U/6SD4HeKvwf8ufL/qf9JB8DvFX4P8Alz5f9T/pIPgd4q/B/wAufL/qf9JB8DvFX4P+XPl/1P8ApIPgd4q/B/y58v8Aqf8ASQfA7xV+D/lz5f8AU/6SD4HeKvwf8ufL/qf9JB8DvFX4P+XPl/1P+kg+B3ir8H/Lny/6n/SQfA7xV+D/AJc+X/U/6SD4HeKvwf8ALny/6n/SQfA7xV+D/lz5f9T/AKSD4HeKvwf8ufL/AKn/AEkHwO8Vfg/5c+X/AFP+kg+B3ir8H/Lny/6n/SQfA7xV+D/lz5f9T/pIPgd4q/B/y58v+p/0kHwO8Vfg/wCXPl/1P+kg+B3ir8H/AC58v+p/0kHwO8Vfg/5c+X/U/wCkg+B3ir8H/Lny/wCp/wBJB8DvFX4P+XPl/wBT/pIPgd4q/B/y58v+p/0kHwO8Vfg/5c+X/U/6SD4HeKvwf8ufL/qf9JB8DvFX4P8Alz5f9T/pIPgd4q/B/wAufL/qf9JB8DvFX4P+XPl/1P8ApIPgd4q/B/y58v8Aqf8ASQfA7xV+D/lz5f8AU/6SD4HeKvwf8ufL/qf9JB8DvFX4P+XPl/1P+kg+B3ir8H/Lny/6n/SQfA7xV+D/AJc+X/U/6SD4HeKvwf8ALny/6n/SQfA7xV+D/lz5f9T/AKSD4HeKvwf8ufL/AKn/AEkHwO8Vfg/5c+X/AFP+kg+B3ir8H/Lny/6n/SQfA7xV+D/lz5f9T/pIPgd4q/B/y58v+p/0kOF4srtfqGtdiVSpQUNV6vXelOlomv1uuxbKEgIKKa7ouBW0ZDQ8ag2joyPbYznCDJk2QbI4znCaZcZyJegAAAAIudAf1Y4d/NHbf8KuwBKMAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEW+Tv5U25+aTp7+9NvEpAAAAAEXOgP6scO/mjtv+FXYAlGAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAACLfJ38qbc/NJ09/em3iUgAAAACLnQH9WOHfzR23/AAq7AEowAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAARb5O/lTbn5pOnv7028SkAB1+2Wys0Ssz90uk/EVWo1WIfz1ksk8/bRcLBQsW3UdyMrKyLtRJqyYsmqSi7lyuoRJJIhjGNjGBwK21NaNdbf7yPL9UGOp8VVO8n2RIWCMYUlKmqx5ZZO0rWV45QiUIE8Ycj8sos6I0y1MVb3vu84yPh1RujUW9qgjsDSuzqHtijru3bBK268tcJb68Z+wMUr5jmWgnr5kR4zyYn1lqdYq6HtkyqmXBy5zwGquk+et6S1tgdL7w1PtiboTsrG6xOu7/V7jI1V2dZdsRGeZwEm/XjDHctXLchnREyGXbrIlNlRI5S5rEXOgP6scO/mjtv8AhV2AJRgAAAAAAAAKWvJ75sNEeO2wVbRVXo1w6q7W2Wg3PrLlLUKTt5a3xX6qaMdKXWZYRE8nTol5k53DBA0ZJ2CWatXTuPhVI5u5kW9f7Dq76Vfd2iVqr/jV4w1/BzftvomoXzdMIpb4SPVUPlm1n8obkbYJIe492ddNVmxcpmN7LlgyXwdsl63fku+kb6BRkZHoXwlUnctajE0VVZnmnf1beyDlHGSfWnLWBg7HuCxqEJhUuPcq1Rqsj7hwupg7QuVyS342+kceOzqOxF1Rsu02Xi3oxo8LFTWjuq4d1rl+1mC5RSWYR95kG6FJdrFdq5aosJeUrtjcZSOtivJoeimb6mjxpINW71g6bvWTtJNw1dtF0nLVygqXB0lm7hEx0lklCZwdNRM5iHLnBimzjOMj6AAAAAAAAAAAAAAAAAAAAAAAAAEW+Tv5U25+aTp7+9NvEpAAUleenlVbpXgvd8xO7r2rRtaaU0PvrY9p07ryTZV+u70tsTUY+Z1oltCbK2Vn3NP17M1+RmSU+JcsGFok5dqpPLqtYVu0cV0+RhZzJeEjxC62kWxnOsd29D+KvUe9SGdqM2OdPWJnGyVhQlzEUTKtEyMzBVyIeoLZMiYkgU6hfRPGcYcmXsry9uj6U1rbl+rx9BoWveS9CbPoVE16wRq9bq+zLpzldkrhNVxrCkZsYOXk27ROSfFYEbqncxiDnGCnTybLjDWGs+bvIH9HyQ5yqFZpDbf3i43RHbxe02JZs19mRURrqh7BiJq9OY5Iik5JNb0q4kkLDLGcPcuzqNvrWUs+5xuQSbtwwjZF80jH026ZMXbttDRikalJS7hs3UWRjI5WZkYiHSfP1CFaNFJaWjI0jhVMz6RZNcKuUoB33aN4uu9OJYmyc4bn1EwR6Quj9OzbDneeJOEdukeM+tEE4Vu31PvnZ9kLJuknK71BV3X20MVtHPCOZZu8PHtH1hQAAAAAAAAin3H1XTeIeSd+dU3o7c8JpjXM/bGsY4XK3zY7Kk2y0qFTbHycmTPbTaHURAtSkNg/vn5TepSlMYtHn0cnhORhNM2DykdTRRbh3F33Mz233t5s5F5GeoenrbILPKXVayeRLnMAzs8PlrYXaUYRAn+nXNWrpDYjYJuhjZvAQX7L8anD/f1bcV/qrnig7LemZYZRl4UjcQey64VMrj6qpXdhwRo+1xuGijpdZFkWTUi1TqqFdsHKKiqR6JU/BT5JuGl3aniG8q92pmtkFVHcNzJ16xzsXWMeUhVjpQ0bYGldtDaJjzqrnTI5i9dRsqVEjYj6TkFEfrI+7HeX0nHndDMHuvxO6D6xMzMszbXrm7bn+lcypWhU0yy0jFO7Db3Bv4kYirpFNrUK0YqayaCsUxXSOmftL3zd+V2ttnzm0/R6Ov1iM2yTnJabanFwXOXPtYVKg1gNdSjt6tjPs+7asEXLv09rJ0C49PXrin0ivs6Kwtmx/R9PJjG+5SIqbOaHsBDCZM+vtHW+uaUR9hPP2ewf/jP2+voObkPpQdRouCPN0eKHyq6uhcGbpuJZ7oCIcEQUXIqf0ySw2mmoGx6JHMj6vEzrJkVPghMp+ybL+vfpVHhyuj9SIs25Nm6XmG2EsPYvcGk71BOWCynp7TZ2pXGVsZJuEcGxlZMjpTGC5wchjlzjObH9SeXTxf7zMyQ1h3tyvYJKQbYdtK+83HTqvalEM5KUxj1K2ycFZkDJmOQqya8UkqgZRMqxCZUJg0863b6ncmCUrULRXbVFrpkVRkq3NRs4wWSUxgyaiTyMcum6iZy5wYhyqZKbGcZxnOM4HYgAAAAAAAAAAAAAAAAAAABFvk7+VNufmk6e/vTbxKQAGJ976aqPRGl9p6IvyswhSdv0Oz66ta1eeN46dSgLbEuoaUUiH7plItmciVo7Vy0crsHiSK2CHO2VLjJMx92r4/edN1cYw3Cex4ixT+lK3RaJSK09zO5YX+uq6yZxiFCusLa41m0/h16rLuHj5ePm2scm3/iSBvfRyrFddip0/k/xn858na73lQ413sjeMj03IPXvQuyejLmrtDZO3m7mvKVJrCW+fcMYxsvX4irLLQUTEso1m3bM3DlRT37tys5PjHirw88r8MbYNuPWtj3rsG1QWtnWltRtt3bUkNjwWgNNPp5OxvdZaajHbBjis1p3Ito8ix3q0vLfUY1qxTkk0FZAr61gRc6A/qxw7+aO2/4VdgCUYAAAAAAAA1PPP7Pyvc3Ynjs8JtDkHP8ADt2bNhOlusFItVQq0RoLXjmcTbMJFZIxCpkkm0VdZxuxMoVZSahaesczZJy3VX2q67X4ep1+Dq1ej20TAVqHjICDi2aRUWkbDw7JGPjWDVEmMESbs2bdFuimTGCkTTKUuMYxgcwAAAAPydMipDJqEIomfHoYhy4OQ2M/84MU2MlNj/8AGcZwMIbG5i5s3Cycxu2ufNI7Pj3fs5dM9g6qotybLmJn2kzqJWGCkSGUTNjB0lM49tM+MHIYpsYzit7bv0ffw6bq+vntfCeoYR1IpYSXf60xZdTO08F//adofW89WEmShf8A+7VJLJsfYf2i5zjMAbh9FX5Upj8ty4T6s7L4X2bHJFJX5/X+35y41uO93nKpW7mElXsNan8c4ckaHfx57+k1dIs00fdJZMdXPT81P6UbwHnP+l7rzl5cNPQfsGxF2lrEae3yeGa4Q97lB49kasZ7JKIYcnOi6tGyZVRYieGCT5Q5WimWNFfSe+U8XJDS/kQ0rvHxn74S9yg+ru9KjZpvXjhwfGUvrUbfomrx79rEOnaEgmzmLHUYSDyViuU00c+Ce82JdYbc1ZuypRt909sakbRpUulhaNtVBs8NbIF4TPrjOEpOEePWmVCGxkiqJlCrIqFMmqQhymLjIYAAAAAAAAAAAAAAAAAAi3yd/Km3PzSdPf3pt4lIAAAAAi50B/Vjh380dt/wq7AEowAAAAAAAetZZJuiq4WOVNFBNRZZQ+fQiaSRcnUObOfswUhC5MbOf+MYzkan3gQij9s97eVHy/W06ksS37tneQOb3LhrgrKK0trjNbk3LmEVUy6IoWUiGOuGjqQjn6rZxJs7FgpEMOVEC7YoAAAAAAAAwpvTm/QPTlOc6/6G01rfdFMdEWIav7HqELa2KBl0zJKLsMSzRwtGu8kNnBHscq1dp59DJrkMUpsa6e1fo0sJpW4SW7/D71/u/wAd231FEneaIzts7d+frMs0L6oMLBX5NV9ZsMHBylTeJTb69w2W+ckJWc59v3nQUfL35b/GS5LW/L9wTIbt0vCqpMVO2OH2B7LBHjyFK2LN3WpP3UfDtHCyiCr9wpJf7THy1ckRaVxd42OR1dzxj5bPHp32xZfDR03r212t0kmotq+ekyUvbDA6hDKZSca9tGYywuip4Kcp3sO1k4vJyHIm/UMQ2MWOAAAAAAAAAAAAAAAAAIt8nfyptz80nT396beMY9z99VPhCDos/bOfurd8Nb0/nGKTfl3SM1ud7V8QLNo9cPrk3h3bXNejnabvBI946NlN0s3dEL7PuDZFV0T9Jt5OnK7W7jF8d+Sp1S7hNRlcq12+FF2jSp+emZROEiYiKt7i3pVt7ISMyqnFMmqEmdVxIqEZpFM4NhPOyGAAACLnQH9WOHfzR23/AAq7AEowAAAAAAAVt+YDpc3IfjM7N301f4jJyr6UskDUnvvkEVW932IZtrilrNTOCnSM7Rs9ri12qRk1MKrJEJ7s+DZwMH/R/wDmNPlLxJcdUNzHlj7RdNcobrvBMplTXNaNxuFb2ZB3giqxTOoiEloSAVUKpkqn8JwoUqZTYTLcoAAAAAAAAAD0uWzZ43WaO26Dpq4TMi4bOUk1266R8ZKdJZFUpk1UzlznBiHKYpsZzjOM4FKPZv0fPxkdmv3dzldIp6G3KZwpIx26ebH5tRXFnNGUKtiXfRlfTTpthemXLhVZ5N1t3JmOZVVvItXKpnArub8D/SJfHwkVDiTyA697+05DZyeK0f2XX0YjYJI1HCBG8Ew2RLS0o9dOUGTNJm0em2ZT4ZRw6cuV4Br9hx9zP6Rr0Lyo6QrPlh8UfTfMThqdNi72/ppJLcmo5d57DUpl2jnKMVFRpV1XGcN2UbsG4mVKX/23RlPaRTsM0P8ASIfDt0Io3ZVbtKhUucXcGaHr+54i26ZkWztNJNVRBV1sWBgINbBMKYJl0xmHbEyxFkSOjqoqkJbhQtpay2pDNbFrHYtF2NAPUiLs5uiW2AtsQ6RVLg6areRgJCQZrJnJnBinTWMXJftxn0HewAAAAAAAAAAAAABFvk7+VNufmk6e/vTbxmzZ1FYbR1rsLWco/kYqM2LR7ZRZGTh1sNpaNYW6BkK+8fxbjJTYQkWbeQUcMlslNhJymkfJc4L6DVYZcF+SPaGmW/ihuHkf8eT3hzWEHrbSFmmdY1qeN3OXVlPlGbPX+ubDXXb5xrWlXKShKgxim9rQdmnVZWIcOWjF66Tdqr7bSCRW6KKBDHMVFJNIplTmUUMVMmCFMoobOTKHzguMnObOTGN6mznOc5HtAAARc6A/qxw7+aO2/wCFXYAlGAAAAAAADVm+lNWKb2PozhzgCovFErP3j2trSkOmjdRI669KocrEP558uwUyUz6MhbHaaTLPCYOkVNZozMZT1yVM+z9Wq9E1GuQFUgGibCCrELFV6FYokImiyiYVihGxzRIiZSJkTbs2yKJCkIUhSkxgpS4xjGObAAAAAAAAAAAB6XDdu7QVbOkEXLZchkl27hIiyCyZ8eh01UlCmTUIbH2GIcuSmx9mcZwIB768U3jd6cSfY3fxRzrdZCRTdJuLHjW0BW7kTL03vHSzS61JtBW1i5WV/wDeO6ZTKDjKvqplT285zmo6+fRUOAW867u/J+2+ueHb6ZZN2wl9B7rkcwzZ41SMVhl1F2xnNTjho0Xzl0RuwtsQtlQ6+CPEsK5yXoafjq+kb8omKryp5Z9Z9Y1GJ+q4Y6/7WpEnideMWhc5Mzzc04jYc29cucETbZy7uUGmVA5jpyDVZIqqn5J5aPO9y2ZNHtvwsvtwVyPw3LLbJ4gtr+1tlEcrejl+SnISO15BDCDUp1FTSElDsTuMELldkgrgyOYdR/SqfFzb5lOobzV6C41vZMezIVfpTTUzDmYqYcZaZOeRoju9IJs1XBFit3MglGnym3XO6bMzJKJluk0l3Txh0iybP9C9U6B2yk7KbKKFI2tTJyS9SHyRRNaIay5pVsskfHsKoOWSSyRvQqiZc5xjMqymKcpTkMU5DFwYpi5wYpimx6lMU2PXGS5xnGcZxnOM4z64+weQAAAAAAAAAAEW+Tv5U25+aTp7+9NvGW9zQ1gsWn9rV+p2dCkWmd1teoatXNytls2qNgk6vKMoazuHGDEyghASK7aVVWwcmUk2hj4MXJfXGgPFcddYw+sLzzPBeLuY0nb75oblWmWjvt30LS43QTHoXSG2LvbpryFP96JTqL23PbizvSVlgK0WRPMmkXSlVIVVsRm2V/odNU1EmzdJZfLpVNBJNVzkpSZcKETKU6+SFzkpMqmxlTJS5yUuTemM5xj1HvFaPlu6n3XxzxTdd1aDia4reGty1rT3l1ukNKWOk6Yp95ucXW7du+6QEOZF5L1fV8M+cWWYaZdMmmGzbLh+6IyQXKeBHHPlD3O30F5O9m752bprrvXnA7M8/rnrLRdU/wBrtWdAppamVv8AYKYyjULDeIJrMUefIwrE3M16xTLDKsyjj2MrN/Rbp/AvkN74+KfijTXdNm0hfa15KuSbl01pdLU2uZXXsvoe2UlnDW9/qmecuZ+yJ36EdUKfQcI217/BXxJxnlnhqojnKznZVEXOgP6scO/mjtv+FXYAlGAAAAAAADVP6qKbqn6UjwFplXCkpS+EOWtkdGWZgdkR1Hx9x2AwnGFfVM4J/wCplJfxcuqpYuXeTEMhFtcNEcKrKLl2sAAAAAAAAAAAAAAAAYc21zvoPfUMtXt3aU1TtyFXxjB43Y9Aq1za4zg+FCnSJYIt/lBUimMKEWQymqQ+MHIcpvtFMe6PoxPht3E/fTjTmeT0zZHrdZAs7onZ+wNe4ZqLHOp9aZVgk5KUVJymoc2SmPVVEzlwRNdNVJNMhIsF+jDrapUKrxt5ZvIbzGkyaKowkD/r81+p8e4UUVUysvW4+Y12g+QNgyRVGuXDfBjJe9Iqmoc2R8afNP0pLkL1S072ty15EaJEZQLH1boKjsdabElGDVuY50s2BNsydFlX65stfantuzDUhEWzn663yo5Sx6jfSAu6+RlUWHlM8OnQGn68y90jNbz5umG25NbJHwh6ru12KCDmAi2plvdlRN/uzJ+8OqZIuMGQNk9qnKHnG8WvZqkZGac661uxuMougyba72m6c6gvp5NwsZulEtIDYiFfNMyKjgmUUkK84mCuT5J9VUXKoQxrYSHIoQiiZyqJnKU5DkNgxDkNjGSmIYuclMU2M4yU2M5xnGcZxn0H6AAAAAAAAEW+Tv5U25+aTp7+9NvGPvIXx7A9r842jU090JvnmRu3byk9naOgdnraxmmrVCAl2MjD3hwozkYez6zkmT1XNyrMyyKk+j23ts5OGeJJyCWr/VPG74wo3ijm+X3l5q/JXtDkPZ9xq/L8TDp9IGgOPdiWulzDyspUNrrtXTks8pGnVp6iuIuvlkLZiKiI5vEZZ3HCZmEufdgQQSbIItkCYTRbpJoIkxk2cESSJhNMmMmzk2cFIXGPU2cmz6euc5z9o9oqM81XNO/+o+Pa/UOe64y2VMULovRe59kaAkp1jW43pvTesrQrN37QD+YlVEIZBtem5mCx2s0uhESpockTIKkbPj5FcnM/js33vmv+ZJa1c4JeP/Qvf2o6dqjQ3Jk3MUxYtPutb1ZP1G27ltFe1FJz1HqR7xY3sK4dtaw7UfyLJgs6k2zhyi1WX+HgDlHu7aHZXj42h1jzDJcy1Hxb8YX7nktlk9g0W2teh9vXeKgNeYnteRtRkZl8y1wzpteWsJ3tkNCypZVzHsCM101HWUdo6TjI6ajZGGl2TWTiZZi7jJSNfIpuWUhHP26jV6yeN1SmScNXbZVVBwioUyaqSh0zlyU2cZgHfec9Cak3pxLZdX6c1tr6wu+kLpCuZqoU+DgJNxEOuM+tHjiNWexrNuuoyXdx7Fyq2MfKR12jdQxcnRJkthQAAAAAAANUrwpLF6r8wPm98gBsIydWj9m0PkXUlhQw5K1fw2skXkLbMtCrGWRWSLHa+1wui+bOPduSyTlVNs3SXKXO1qAAAAAAAAAAAAAAAAAAPwommsmdJZMiqSpDJqJKEKdNQh8ZKchyGxkpyGLnODFNjODYznGcZxkVO9eeDrxfdrsphXbnKGuIG7SyZv8A9WNRxDLVG0Gjv0zhGQzZqe2jyTbpsbPtoJ2phYGJTY9DMzlyYuagVuCPOl4lljS/jh6dZeRLlqEP/wDbch9dyCy20K9AomPkkVQ7kV/AMV1EGpjEb5g7RTWSb1JssnQZZJVy1E0uKvpFHKG/dg45v6yqF48ePXsc5RiZXTHTLU9Vg5aaVPhNJrUL7Ktolkud5nJcsmFrj6rIvFDZRi0pYpSOFtg1FZFwik4bqprt10yLILoqFVRWRVLg6SqSpMmIomoQxTkUIbJTlzgxc5xnGR7AAAAAAAEW+Tv5U25+aTp7+9NvHaOh5fTUrr+06T23tSl63JvjX2zqHGIWK6V6qWCZh3VJlUbs+qTWak451LOKpWXjuelVo0q2IZg3zJSB2rRMy5dPKbhtzbJ5ZrPio2H5YvCDF8C1usa91I+3rUt3UJfpWzaR1rMRZoWJQoExeVtfVzYsnX6zBNJK2tpZyo0lV3ssydqSKZXy28G2bptGzdqj7XumyCTdL2zZOf3aKZUye0c2cmOb2S49oxs5ybPrnOfXI9wAACLnQH9WOHfzR23/AAq7AEowAAAAAAEHfJZ040424I6x6UWfJsJLV+k7tKVI5zEKZ1f5KKWgdfR6PvE1iZXkLpKwTNPJ0lUyZWyqsTKKanpAP6NvzE85n8SfOmbG1ylf9/EsXS99eKoFQdyEht+UPM1ZV0Uq7gntt9do05p6pmSTVOio5w2bKOFUi3vAAAAAAAAAAAAAAAAAAAACDHb/AI2uMPIlRs0bq/SFY2Hlm2cpVm6okVr+yqS4cE9MPKhfoQ7OxxBk1SpOFI7664g5BRBIktFSDfBkDa+6fE/mu8Lq6sp489tKeSriOFOo8Pxzvt82jtzUKASNkqcNrK5qqLunyMWwIkhHo1h4waHwXGCasfqI4UVsX4X+kF8P9fWhLSuy3Vl4s6uZu04ee5y6gaHoE6WwmUKhmIq1rnG0PA2V0q5Nhuzi3JYG0u1Ml9islKdMx71iHIqQiiZyqJqFKdNQhsHIch8YMQ5DFzkpimLnBimLnODYzjOM5xkfoAAAAAEW+Tv5U25+aTp7+9NvESvKDcL1rSG1Xs6g+NCv+QuSpiO0V5aUmdj6c1kvoSrSlOxF22eQntuJmSxG3SsOJeCsKcIqiY8M0coyvvGS2CZ1m223tH50LUOp5P6JpztW+cLq1qMtDbnsdk5Siqe3rN5kY2Nq13msK6zNOQNDlVZeMeIXaWhmlZLEPms4pKEh1iPjb3mM4NjBi5xkucYzjOM4zjOM49cZxnH2ZxnH24zj7M4HkAABFzoD+rHDv5o7b/hV2AJRgAAAAAANXf6VvZpOxcS808j1dw5za+1O5NGaaJGx7wjWUeVZtmdnJZdkVU3uV8oWpOhsjpu01WRTSiR3BcmKkU+y/RqnF0Kk0+jQbZFnC0yrV+pw7NumRJBpF12JaQ8e2QSSImmmig0ZopJppkIQhClKQhS4xjHaQAAAAAAAAAAAAAAAAAAAAAV6d0+LDhryMVdWC6i0ZWbRYUmijWv7WgWzer7gqJjJmIkrXNhxjcs0kk3Pkq5IiUPKV9ddJE72IdYSIXFEq/CPnW8SSx5nx19LtvI7ypBY/wDtuROspNZXa9crzY5vdQuvbirIQrNQ7WOKm3ZfwK0VaPI8IQ6OupBNVVuaUnMv0mPi252c+le46tsfxu9KwuU2Vm130rX5WIp2ZL2vdnUhdhYjUGzWMXUKoZu6ukTUks4TOmk4d4wku42BNb7V1luKstLpqbYdJ2ZUH+PVnZqFaIW2wTjOPXGSpykE9fM8qFzjODp++woQ2MlOXGcZwO+gAAACLfJ38qbc/NJ09/em3jPF9qFb2DRrpQrm1K9p93qdjqFrZHcKMyO63ZYd5DTrU7tJRJVqVxFvXSRnCSqaiGD5VIoQxcGxpz0jl7knabef4sl/pB9+2z49tJ23V+mp/jvNf1tWo5q2mLZKw+pdC2DrRl9VlbhTnbunOqWwg4ouVFyV8jXEqzyg1IlugIopN0kkESFSRQTIikmTHoVNJMuCJkLj/wAFIQuClx/4xjGB7AAAEXOgP6scO/mjtv8AhV2AJRgAAAAAANVHyh5J0d9IW8LPKhfq0lDaIg9g9m2+LWQVVTbuod/LuaW7OqUvufftZvUCTjDYx8+yRZsq5J7tduVbauAAAAAAAAAAAAAAAAAAAAAAABGro7jflTryvFq/TfP2qd2xCSCzZni/0+JmpOKRcZKZbEJPqNyz0EdQxCHMrDyTFXJyEP7ftkLnFCWyfouHMtTtD7Zfjq6d6i8cWzFsqrtFtT7FtN3oH1kxnKiaEjWbNaGNqexRTulfZjM7AIyQxnGGzdIhcpH6lnSf0qnkzKp9d9S8heRqlRJnSjCtbhp8LqfZEiwbI4+qMkpmNhqQgvIPMkzlVax7KkzpOlPYzKKNs4UR8k86nlP0Q2UjexfAt05mQiTpM5S58wT0htiqSS6aWMO5GPiYqn2RKNjDr4yduc94mm5ET4IpImOn7avca79LF8dDOScQm+tU9n8xy8flFOWJtXnyTVYxblXCZjNXhatMTdhRWTIoVb2Vq0jhRHODJ+0cxEzWic7eZrxcdTmYtdO9uaHkZyR9yVlVLfcGusrg6WXbmdFatKtsgtVm3zsiCah12rFm6Wb+6VwsQmUzellkfJR0s0RfxT9lJsXBCqN3se6QetFyGxgxTouWyiiKpDFzgxTEObGcZxnGfTOB9oCLfJ38qbc/NJ09/em3jv3RrSvv+et8MbbZX9MqrzTO0GlmuEWgu6k6pX3FInEZmyxzVqU7py/go47mUZoNiHXWcNU00SmUMXGdLPR3jY8kXUnPdU0dR6FwETiHe/IPMmhYDsauEucTZZvSGobrZthVPf1b0nPV+JtUT0rdYS4qqyMpY2rVm1t2FbAWQIR6ZyTesbIFatm7Yp1VSt0EkCqLn94soVJMqeDrKZxj3ip8F9pQ/pj2j5yb0x6j3DjpiYia9EyU9PSkfCQcMwdykvMSz1vHRcVGMEDuX0hIyDxRFqyYs2ySjh06cqpoN0UzqqqEIUxsdI1huPUm7YBzatObOoO1KyzknEO7sGvLbBXGGayzQiSrmMcyMA+ftEH6CS6KqrVVUi5Elklck9hQhjfDrre+k9vyVoh9Ubd1psqWpDwsdcYyiXet2x/Vnx1V0CNJ9pByT5eJXMs1colTfERMZVuunjGTpHKXKwi50B/Vjh380dt/wq7AEowAAAAAAGqnxKXPSP0nvyeb2XK/ka5yHzRrrlqquV85eRMfaLO7o8xPYjVlPeEjXjJxW73HrMmhkMnPLS7hwTJ3B8n2rAAAAAAAAAAAAAAAAAAAAAAAAAAB1i0UmmXdirF3So1i3RixMprR1ngIqfYqp59cZIo0lWjtucmcZz6lMnnH25+z7RVb0V4GPEj06aQebC4j05X7BJfXDurZqKENpixrun3u8uJB281ktWW0rI5MkQ5Hsyykl0z5VyQ+PfuMK9s8cPiQ578Xj/bvw8bL6OslV21ms4zr3ce0kb3SqAnWCyJWxKBFoVyCWjVHpZIyT97Lu5ySVbNWjUj1NBIxFLTQEW+Tv5U25+aTp7+9NvGeb5Cxtjo1zr0zMO69Dz1UsULKz7B+nFPoONlIh4xfTDKUWwZGNdxjVdV62fq4ymzWQI4PjJU84GjjrXSPgAZbQYc3ax8x3kEqDtW1uKxAmrvTW3KLop5b5GbetV4aubGaUGG1Aq8kZpNxgmYydKzfLOEVm66xHaCim9yA12/pNtxu9f8AH7qunVGuO7fC7q7l5Y0/s6rN7IaotbRrewWGenZOoTNiK3dFiYO72CsVqlSj1ZE7dq3sOXDlJygko0cQcg7Ovyhqfzw6lguRNRcq9pa843rG5Zd5yPsTYrzQextd2TWF7gNYWGm0qyLR8ZrfYdE/h0/F3GSgarCPLbJpJWFZyuXKJGvUeJNT6v5Z8gfgBbc21CuURv0r4vtzMt/u6iwSbKbVbQmvaLseHsV5Va5yacnGl4Xcv0bLKGXkTGOdhl5lqUrcu5PJrP28bIuIpkhJSiDF2tGxzl7mNbSD9JuodmycSOGr7LBB04Kmgq9wyeZapqGXw1ce790eAd9te8pzenErTZWnafr+BJ0hdHDaagNwKX964lScZ9aJt41SFPreoYbtl2qr1yeS/ia2UFGiLb6kr9bys3sKAAAAAAHHTEqyg4mUm5FYjePh459Kv3CpsETQZR7ZV26WUOb/ANJCJIIqHObP2FKXOc/Zgat/0WiLd7S1T5De85psXEr2x3xtW3QshnLjCj+i1ZT65DqmQXzgjY2LNdrk3UKmXKiuGyOV1T4TRSb7UIAAAAAAAAAAAAAAAAAAAAAAAAAAAAAIt8nfyptz80nT396beMq7uqKuwNL7doaEWabXu2sL9UUYUkwavHl1bJVJaGTiyT5UljQZpA70rQswVFU0ZlbD3CSmUPYzp43PjPzd3fxx1Xxpy/ja4L/2nrVS17rxW9NN8QjK+yFQ11Mwj1jMtMlgV4mB21Nx8C1JM7NZN3Dv/UT2VtTGHbP3SKLbaL4dvHbN51nYnnc2g9U89bFi7eeIp9V1JtJ1tiEmKGhAQarSfk552yYqsphScWm41SNymp6NI9o8yrkzvJSzSEaeu+S9LdvaDu3OO/IJ7Na+uyTFZReGkTwtnrFhhXqMpWblT51JNZaCttWmGzWWgpVNFYrZ43Jhdu6aqLtlo1creLLnzmSE36hM23c/Ud56irsXSd87d6tvTbaezNha+r9fkatXNeSUy2gq5Ht6bBQEtJMGMUzikDny8WcvHLpxhFRLofE3hy5s4b2423NS9kdH7fstR1c80bpCN6B2kjsSuc5aXkJ1Kwv9baVjEa5BHrsE7eNY9qs5lHM5L4io5tGoyCTdaRw/tnEXOgP6scO/mjtv+FXYAlGAAAAAAKrPN50IbmDxRdxbabPyxssjpGboVddfWVmixLJt57H6ogstHDYxHKT0klc26rVRucqqSyZVSnJhPJy/F4Nee/hj8TvD+sHLXLWcX0nA7FtRFPcmWxaNtrO9mTCCqiCSJFf4cvasRDc+SGU+px7Yqqy6hTLqWwAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAi3yd/Km3PzSdPf3pt4ksrJxqMg1iVpBilKPm7p2yjVXbdOQeNWRkCPHLVmZTDhw3aHdNiOlkkzpoGcIFVMTKyeDVa9UeWjVfLW1di6yc8+dRbpaaOq1buXQmxNJ0WoWKjaSgbVCObVGK21xYb9VZyQfJ1Fqe1vmFYhJtZvCqNzJZcSCxY/Fo8PLR8/ERU7EuCu4qajWMtGOylOQrqPkmqTxk4KRQpFClWbLJq4KcpTlwbGDFwbGcY5EAABFzoD+rHDv5o7b/AIVdgCUYAAAAAA1YvpSUzJ7T1RwD4/q0+WRsPcvb2uqw+ZMnSJHi1H1++ilbFIHYq+v15hDTN0qkmsRQuWybxqxUUwZUqBc7RULDxtdhomvw7VNjEQUYwh4pkjj2UmcbGNUmTFqkXH2YTbtUEkiY/wDBSYwOTAAAAAAAAAAAAAAAAAAAAAAAAAAAAABFvk7+VNufmk6e/vTbxX75e73pmJxoHXli5o2P0H0DdnOxprRcppzfWu+Utp6qY0tjW3F8uFO6I2DsfV6NWfkzL1fH+lYefkXdrwhhd7X5CKg3qjatvXvNNi6op2qu2rnyxQvJ5Qk0pynv6xup/Qdad10w2qLbP059SbTsHX9nR5O7SgqnZoScZwby1FqTSfZqGlIuxT7KdPgu1tCOivYWIekjHUKR3Fx7okO+bps3sSVw0SVLGPGiRjpNXTDB8NXDdM500FkjpEMYpcZzygAACLnQH9WOHfzR23/CrsASjAAAAAAGqZ0Wnjrb6U5xrqdXGJmj+PHlG675tEaqRs9jYzYmy4yTQrhlUFDGywmUXM5qmyoOcpZXynERZ0cJ5wm4JtZgAAAAAAAAAAAAAAAAAAAAAAAAAAAAACLfJ38qbc/NJ09/em3jue+eZueuo6q0o/RuldZ7uqMfIkl4+v7Np8Jb42PlEy+wWQYITLR1hk7yn/7R12uUlFEs5SUMZPOSjJNKpNO1vVIGia+q1fpNLq0chEVuqVWIYwNegotqX2W8fExEag2YsGiWM59hBsgmTGcmN6e0Y2c9oAAABFzoD+rHDv5o7b/hV2AJRgAAAAADVM8I/s9H+XzztdvO/dSLGO3XSuTdeS+EVEjYh9XKz0HamqWDYwmo3+qUHXX1d0XJzuUyGWx7CSpfe7WYAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAi3yd/Km3PzSdPf3pt4hr5gaf5JHvPiGzPGpu17Qtn6mPKWW46ljNbUy/z286dgrFWTgaUpb46UTYbBg2LN+9qESg1yjbZByaBWVbuXDBYkf6XxL5eLjUKnb/8A60Wwq3i01uCsea3bOCtGQFpgcTcY1k8wlkg37tF9Cz8X9Z+oTEW8RTdR8gg4arplVRMXF9wDpGy9j0rT2u7ztfZFgYVTX+t6nP3e6WWTU90wgqxWIxzMTUo6NjBj5SZx7RdbKaZTrKmLhJFNRU5CGpS4B8y7zqzlfv8A7C2tqFfVGtuRtj7LSqVSMnKMb5PakqGuozZNZlLg0nzIpxd0tEFKMllI9Fu0YR6j5uh6K4TMup1fgryu9Wba6L5m0p2joTTuoY3vPmWy9Q8my2prvZLPJx8RUcxkvM6w2u0sMe2b/wCsS0icjrOjL15QkNnCLhhlIzpbCLPYMEXOgP6scO/mjtv+FXYAlGAAAAADB3TW6oLm/nTe3QFlWRQgdLaj2Fs+TOup7pMzak1WUsOUMn9hTODOTsCNk8FSVOZRUhSJqHMUhqLfosGo56leLGG3LcmzpK69c7z3N0XNOH6qLmQdx9gtB6rX3Th4nkyjssnH1PNhRVcn+sZxOnyqmipkyRNkMAAAAAAAAAAAAAAAAAAAAAAAAAAAAAARb5O/lTbn5pOnv7028YC8jnLk3v2lUq7VPvTbvj7t2ln0/YITblAslTjaQ8NLsWqKrDbtYu5mtaulTaqx7Z4WJkpWLSwYrnB3OUl1SZ1smPn17X5X3jX+c4i581+dRm9kV4tSf4SqWzIrfcMi3KqZB9aENe0y9aGnVHDbDb3ydPmFzkfYksLqtWzJRQm7cAq+8sHHHQfcWgKJpzQmzNYUAsXvHXmy9mQe4K9ZLLr3a9H164fWFprK2RVTexsy9rkvdG1VlZ2NSkWTeXj4VWMeKmbOlUlKRuQeDfI/srT3nh0F0az17CQ3Ttq3rXqbiM1PsbUrTZ26rFrmmxkBsvVU9c1lot1z9Ix7E9bZHYJS50Z1Bdf+NKN2i7YvKePuldH9QdzeLm6Xrk3pPmiA8XHCeytN7zsm+9VzGu6xc943OuVrVjGm6dmJhTCWx4JJhAPrgrboBNaJTiyt0V1EVZGOy522JOOby8bIxLtR8k1k2LuOcqxknJQkkk3et1GyykdMwzthMRD4iapjNJOJfMpJg4wm7Yu2zpJJYkA77oaj6w3pxLYK3O7nk37npC6RKjfYfR/Q+3YQrVxxn1o7UVb1nbG0brW2kmVVigRCaaRKMy1bHeMmz9FnIyCDmwoAAAAAGtb9Ka3XYKl44oPmTXxl3W1O8Og9U8yVKGYFVcSTyLl5VS1WJwRk0UxIOWKi1dg6y7IxbPVVHFqYNDtVEnRzFvi5l0pAc3c66M0BV0sJQOmdTUHWkdn7MncJ06sRsGq+XNjBcKOpFwyWfu1cFL7105WU9nHt+mM4AAAAAAAAAAAAAAAAAAAAAAAAAAAAAACLfJ38qbc/NJ09/em3iEPko0lwh070Fwdzj2nq677akdn23cyumazHWOXhdZYnaZRGNttLnaUfCT8E7nmeISLSTgGLgkmwxIHWy5ZY9sixMdr98ePHgvx/dJdac8c3GqeiuSd72fmy/wCvdN6uoGtrJJ7FpG5YHSNnUhmmZCFipuLbWywov05+Zl0XstGIuXhk8vDkQUvGAAABFzoD+rHDv5o7b/hV2AJRgAAAAA1S/IQcvW30kHxU8ptcYmalx9rC59obFY4SbPmEbZzSL5WmGlETYMdk6jJKl0F+1XX/APQc9oiioFysoU+NrQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABFvk7+VNufmk6e/vTbxX55UebO49pbw4R39wXIcsl29zFa92y2a/1TZ73BVKwMtl69aUpROLYa9hJOfmXjBmpIO1U03sUm0OVouoo6TyogKI9i+Ir6QJs3i3qDhmdceLlpqvrDo699N7BsMXsDotK8w932DueE3jMRNTfuddOoRnWm1pgWkfHtZSClpFOEUXbuJRw8MR8nvBgAAAi50B/Vjh380dt/wq7AEowAAAADOcYxnOc+mMfbnOf+MY/wC8jVH8OShOvPNR5mfIajhOTpVGsVd4d0/Plwk4aOYWkuYRW4ZiH6CqqCyL5XWtRn1ip5yoihYmRTHxhYxBtcAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAIt8nfyptz80nT396beKmfMXu15y91/4rOl5PU++9sa41PeunDbAi+f9Y2PaVoYoW7TjarwCzuCgSlSI3UmZBH2FZB02TwVJwohlVVLJM1k+ODzZY55x24XpHmHylXk26vID0lvnR5j8wbWvH+kufNjTEO71pS8ZsEgn/pMkAzaPUsUyH95BQPvcpx5zEVON0AAAAEXOgP6scO/mjtv+FXYAlGAAAAAhd5FOqK9xPw/010/Y3rdklqnVFjlIHDk+SEkLzLoFrevIUvs4Mc6s3eZmvxKZEymPkzzHs4+z1xW19Gi5lmOcvEzo6XuLB0z2R0pL2zp/YC0mmXEzIv8AaT9LNWkZlbKSbld+917DU92r9eyo8b/WMNF8kOhlIl+wAAAAAAAAAAAAAAAAAAAAAAAAAAAAAACLfJ38qbc/NJ09/em3iUgAA4efsMBVIWTslpm4it16FZrSExPT0kzh4aKYNyZUcPZKTkFm7Ji0QJjJ1nDldJFMuMmOfGMeo69r/Z2ttsQZrPq3YFK2RWyO1WB7BQ7TB26FI/QTSVWZHlIB9IMiu0Ul0VFWxl8LETWSOYmCqEzn5KRt7U+zH1hi9c7N19fpKou8R9qj6Xcq7aHtafGOqmVnPtYSRfLxDoyiC6eEJAjdXJ0VSYL7SZ8FyIIudAf1Y4d/NHbf8KuwBKMAAAABqcfSCJmc7j698cvhT14/fZZ762Wl0R1OeKVzg8Jo3WJnkhHJSBkzGI3SeNIS+zSJnxCpYmoSplaldOniSONrGBgoerwcNWq9HNIeAr0VHwcJEsESt2MXERTRJjHR7NAmMERas2aCLdBIuMFTSTKXH2YHLAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAIt8nfyptz80nT396beJSAACCPkc595r6K5hnq/1/fp7X/N2urPV927XcRtqjalXLPWtVPFbKtTdmO5KJlf4trWaXRQPZa4wzGyc4dkwZNJJLJjt3WtxzjDWjXnHfnc8lPJ2nJfl/lfojRx3HEej4aJxr9R9G6Z1Ja6vOdMV6nxOUkaOhtZ/KpT9dSbxUfIrs4ckkZVyzNFuR+OG9Raw5G8gfgSZc3VSDoaHV/jH3Il0c5q7f6rncDmA1/SNmw1vvn1ZTJbBaWdzdPHaVplfrUnlFRSOw7w0KVEm5VJuHrWNkXUaw/isi2Yu3DCL+tIsf4k9RbqKNWH11xgyDP64uVNv9aXLlFv7z3qmMkIbAgHfbvt+yb04lYX3R+dbQyfSF0dIT2dlVa4fWZFPjPrRJCK/hcK2Rdo/WG6zt19eOfKCX1L3By5UcpZxYUAAAAOGsVhg6jX5y1WaVZQdcrURIz0/NSS5GsdEw0Q0Wfyck+cqZwmg0ZM26zlwqfOCppJnNn7MDVU8E8NK97d9+RjzU3OLdq0rYtzzy9xo5l0D4O103rdVOLsU9DpOTe2wRmGMHSmyxWbduiScc3dFdV28XenLtiAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAi3yd/Km3PzSdPf3pt4lIAAIB+SPx6a38nHOKnL+3Npbp1Zrx9d63dbA70dO1CvWK0f6WSkzxlYnndzpF8jnFWzKvmc+5Yt4pq7WmIGEW+vkbN3DV1jTkTxZU3limbl1tbOr+0uxtbbq1/F6wl9f9ibghNo1So0xhHz8O+iaBGQNGpJa0lYYaeNEzmE1HJV2MVDJNCM8scGU6jxD4btCcP7gjN0QO4ukN42GgamfaE0BEb+vsJcK5zhpWTnEZ+Q19qZhDVOtLM2LxwzjmCsrYXU9NFh45CMRfJILPvrluoi50B/Vjh380dt/wq7AEowET9495cScy2tpROjOuubNE3Z/EN7AxqO3d16715ZXcE7cOWrWYbwlrsMVIrRjlyzdoIPiNzNlVmy6aahjJHwXELPy5eK+QdtWDDyO8PPHr5wi0ZtG3Uel1nDp05UKi3boJEuRjqrLKnImmmTGTHOYpS4znOMCX1C3fpjaiGHWsdt6z2I3zkpffUe91e1p+0cuDkLk0FKPsYMYucGLjPpnJc4zjHoMnHORMh1FDlTTIUxznObBSEIXGcmMcxs4KUpcYzkxs5xjGMZznPoIJ9P+TrgHjaLkZDozrHSuvH8e2Wc4qC93h5rYMjlFEq/1aI1/X3Epb5N0oRRL3abWHOXGFkjqqJJH95jWtteyvIJ9Jhlj610pWLrwx4b8WFoXYm47ozcQO8usq5HOMYkK1T2DdSSjHFZlSqfYwYrr05ms3Ora7NY5FnilE20ucOdtR8m6Q1zzvoqqNaXqvVldbVuqQLY51jptkjqOHkhIO1cmXkJiZkV3cvMyTgxl5CUeu3aufbWzjGbgAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAARb5O/lTbn5pOnv7028SkAAAAARc6A/qxw7+aO2/4VdgCUYCvzpzxWePTszYjfbXT3Ker9xbIbV6PqiVxtLKT/AIzmuxLh66jIpVeNk2BV2zFeRenbe/IoolhwchT+79kpY2uvo9/hmdtl2qnAWlkiLpmSMo1xbGbkhT49MmRdNrKk4QUx/wDxVRUIoTP2lNjIh7sD6J54h7Pl0417Td86CkHLk7vLrU2/r1lEqxznPkuGGw3F+bJti5P6JoNCtMIkTTTQMkmXJDY4V+io86WM7uJ2h315Hdm0AijfMBQp/fTc8VAotyl9EEjva/Kt3Popj3iR/wCHNsofZghfsxkTS5v+jh+Ibmx+0sEby5G7kuLUySprh0ZYpvc0g6XSbna4cLQFmcZoaah01D5OVnUGqWVMkPhLGUUPdXdQ0LD1yJjoGvRMZBQcOzbx0TDQzBrFxMXHtEyotWMdHMUkGbJm2RIRJu2bIpIIplKRMhSlxjHJAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAACLfJ38qbc/NJ09/em3iUgAAAACLnQH9WOHfzR23/AAq7AEowAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAARb5O/lTbn5pOnv7028SkAAAAARc6A/qxw7+aO2/4VdgCUYAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAIt8nfyptz80nT396beJSAAAMS773HV+eNJba3vdiOlKjp7Xdv2RYkGGEcv3MTToJ7OvGjHDhRFDL14kyM2aYWVTSy4VTwocpPU2KXODPLR0turobnLSfYvN+rtFsu5ua7H1DyHNax2POXV84rNTUjZGb11tpjYIeHLG39KnTUda0nFZy8gztSLtDGK7PlJrf6IudAf1Y4d/NHbf8KuwBKMAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEW+Tv5U25+aTp7+9NvEpAAAEN/IboOf6l4W6152qmCmtW49A7ModYSO7SYJuLHN1eQbwTRV6v/wCw0Sdyv1Rsq5WzhJFNUyihikKY2NdPx8we8ur+6/Fbb5zmbobQEN4vODtm6j6HnN36rn9e12V3jc61WNVR2vtZy8+Volf2pG8BI25S0VtN/DEiECJrLoLvWfvduaUj0JaNkYpyq+QbSbF3HuFoySkIaSRQet1GyqsdMRLllKxT5NNUx2klGPGkgxcYTdMnKDlJNUlcu6OOqcvsjkdWIn+p5SPZdC2VzaHxOs+tZgtcgj8pdOM28ytIqbfdnqqa9md12vYnm68Ys4WnUquZ8ohZF42QkL8Iesvvx1H86XXf72h8Iesvvx1H86XXf72h8Iesvvx1H86XXf72h8Iesvvx1H86XXf72h8Iesvvx1H86XXf72jC8FzBGuOiNpVmRuvWmNYxGl9CTtQUV6462QiT3yxXjpCP2ORlYsbeTXlJFOvVfVhpSFUk3aUC2UiHyLGOPZF3Epmj4Q9ZffjqP50uu/3tD4Q9ZffjqP50uu/3tD4Q9ZffjqP50uu/3tD4Q9ZffjqP50uu/wB7Q+EPWX346j+dLrv97RhfnTmCNs2i9VWDaF160T2HL0uGfXEkt1x1tXZMs8u3wZ9h9BIbejEYpxhX194yTj2ZEc/+kqCePsGaPhD1l9+Oo/nS67/e0PhD1l9+Oo/nS67/AHtD4Q9ZffjqP50uu/3tD4Q9ZffjqP50uu/3tGF+kOYI2r8777s2rLr1ops6u6X2lO64TiOuOtrJLKXyIo87IVAkXXV9vSiE9InsLeOKyhVoyRSlHOUmKjF2RfLdTNHwh6y+/HUfzpdd/vaHwh6y+/HUfzpdd/vaHwh6y+/HUfzpdd/vaHwh6y+/HUfzpdd/vaHwh6y+/HUfzpdd/vaML7S5gjYi8c3x9QuvWh69aN0TsFtMzLrjraUbJ0NvzvvuzRx5d8pt52etx2dnV3XCaU0g4i1V5ZSLruXyiE8tGSOaPhD1l9+Oo/nS67/e0PhD1l9+Oo/nS67/AHtD4Q9ZffjqP50uu/3tD4Q9ZffjqP50uu/3tD4Q9ZffjqP50uu/3tGF4LmCNcdEbSrMjdetMaxiNL6Enagor1x1shEnvlivHSEfscjKxY28mvKSKdeq+rDSkKpJu0oFspEPkWMceyLuJTNHwh6y+/HUfzpdd/vaHwh6y+/HUfzpdd/vaHwh6y+/HUfzpdd/vaHwh6y+/HUfzpdd/vaHwh6y+/HUfzpdd/vaML89cwRtl1FVZrZl160TuzxezlmCSvXHW1ef4TaW6eZRPv4hHb0ak1xmEbRxkTFZI/WkMpOzZVMvlZTNHwh6y+/HUfzpdd/vaHwh6y+/HUfzpdd/vaHwh6y+/HUfzpdd/vaHwh6y+/HUfzpdd/vaMN9FcvxdZ593rZNX3XrNTZdf03s6c12nE9c9bWKVUvMTSZx/UiRlfX2/JoTsgafbx5WUMtGyCUm5ymyUYuyL5QUzAjyJrQyKRlLx1J7ZkyZP69oddlz7eS4yb1x/vZj0z6+vrj0x6f8AHoPZ8Iesvvx1H86XXf72h8Iesvvx1H86XXf72h8Iesvvx1H86XXf72h8Iesvvx1H86XXf72jC+1uYI2HtnPjOm3XrQ8NZN0OoLZmWPXHW0qgnQyaV3JOonlXam3nhq9Hf69g6KQs0itGqmkTx8Ll9lKZVYPs0fCHrL78dR/Ol13+9ofCHrL78dR/Ol13+9ofCHrL78dR/Ol13+9ofCHrL78dR/Ol13+9ofCHrL78dR/Ol13+9owvVeYI15vTcVfmLr1pjXkJS9OPqOov1x1s1jTz0242aW8lYTpdvIrTDgiUVUv4gyVkHpIfBmZkkGOZNTLvNHwh6y+/HUfzpdd/vaHwh6y+/HUfzpdd/vaHwh6y+/HUfzpdd/vaHwh6y+/HUfzpdd/vaHwh6y+/HUfzpdd/vaPh45objW2uL7U1WtxbtmnRPRj2IWvk1bbNZZODltwWyTiJh3ZrzIS1osbeUZOknkfNScpImfsVW6rd2q29znEsQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAa5cN5172/wB1xNikeTo2N8e1k7nm/HtXups7iam2E43XEPnFabXB3p5xU2rNPVs1eWy1aZSre6LySSKKsmq2UVLiLzsaAAClfpTzGUfVHkG5g8feq9ezmzbltDesFqPfmxHcTZYfXOkFrLr6S2DXas2tJowsLaNrT0CgysSFVZPlUIuqnWk5Ncjhdq1HN97eSfe2g+jKVyPx9y3AdRb3f6E2D1DsOOum28afq1M0/QnycS3RYTBKlblp27XSaK9ia9EqN41igu3QWevTIOT5bTm4j6xonc3KOjestbMnsVUd20hpa2cHJOGzqTrkkm6dw9irMk5Z5+rLvq5Y4yVhXS6OCEXVYmWKklhTCZZTgOvW+xt6dU7PbncfMyzWrV2asbmKrkU9nbDJt4ONcya0fBQcai4kZmZeptTNouKYILvZF8qg0aoqrrJkNQxoXzAdVTXRHHOuut+BMcvaq8gj3ZkRzTNOtvPLHuqvTlBiDWZhDb603K68qjugPLDWUFH6ybeZfOYB66Yxz9oofLxVr1Tc3nVvurtxbzskZyZHWLgDlnqjX3HvQPTzncLaGv8ABbNu76Dr0nZqtqNapLMpug0S32eBgbC8cXNnJOivcPWbZIpFEUtjIpinKU5DYMU2MGKYucGKYpseuDFzj1xnGcZxnGcZzjOM+uB5AQ+7W31vXQOq4SZ5t5onepdw3XYFV19VqKzsJKXT69/qJyoR/fto3k0XOnqOvqszQVcy8o2hJR0ddVo0RbEwuo4Qgjojy/pz/HXefRvSuksagvvjkv8AtfWW/KDRL022bTrJadaVyJsbTOrdiZhq+jOM7SWdi4lulIRDN9ESq+EpBHKaiJj+eB/KRu3oLoeO5c675Yr/AC5tbZXMlV7B0OjUdwpbah7vpqxyKEbIxljUcVOmyFX2DUXEjEkmYhNlJMlsLvTpuWpWaWHd1wAIad99p674B5b2V0rsJq6nv9KR6cbRaDEmUNZNo7LnT5j6Nrqstm7d48cStmm1EG6qjNi+UjIskjMKtVW0cvgQNsfmAkq34qOYvICnohKy7f63Np6naf52iLsdlFS2392za0NVak82C+gFFmcEwI2dyMrPZrSjj6oyWwjGFOoX3chfHj3te+sLT1LoTf8ApaF5+6u4zv8ATqZuzXFV2GltCluIrZ1VUu+sbtTLhmDrT9/B2qtJOFTN5GCZuWDxosgc6uTeynZwACgjqjyw9i0juvbHE/GXCmrupZTS1G0zZL3Zr12Zrnm9+3sW6Yqdnq5Uoeu7BryhLC//AILDlkMEg5aQfe4clWdRzNA7VV1mTtTyRdGaR3Lofk7mLkyt9AdY7Q0Ncukr5Q7fudDWtK1prbX6DdlNtSXRnUrZ/qSyTVxUcVGrIpMI6MeOm+Hzl8m3WKQs2ODOwaP3xyNpLrfXkU+r9a3HWXspmtSbtq/kqtYa/YJinXCrvnrPBW7txXrbXZuIy7TSQw7IzI6+rt/fe5JLsAFB2mPLv0l0Xu+HktG8K52Nw1K9OWPmEm+IPcrFXdXvqfJvazbN8l0MpUkif7E1q3xr6Ik54tyVkMM0DSCaH1rJIlTh+2fNHt3nPeHVNT0dyAw37pHx+UPVWwu1dmyO4m2v7PWo7aBlJLEJqOmL1KaaXqcqdOT/ANVThZSwQDc6ZV45L3SqSThze9RLrXdk0inbEqD9OVqd8q1fuVZk0slynIV+zxLSah3pPZMcvsuo963WxjBjYx7fp659PUdrARc7A2R1FqrTbm2cg82QnVu407HBR7bUtg3FWtGRziuPVHBZ2wGvdsZP4hJaFTTbnSijN8uJHLjJUDlykb1rL4N8qvUvQl+7ar/WXEVW5ZpPDVYXcbS2BTOka30BDZ2DH1wl5nNbtn1Yq0LAuZyt0b2puy/wqclc114qxhZxGPkXZEcdR4k80G2eh958s0Xf3I0Xz5qfyDULamyOGtjRm5G1+sF2hNVorT76E2nTlalWs0qyTtBw2ukSlESlgZ/V3STD3i5veO09gUAFWvkB7z3DzRt3lHmbmzQ1P3j0D1q+20tUUdp7Y/2X1XV63pWos7fbns/cG9Vuco7m37R+2aV+Bi4JVRwcrp26coNmxvbjmt5qWiHiyrHkBNz5Jq7Vu20yc3U7nJteWTphZOj3m2JXTUVVYjaCEIdnIUiQs8S6lUrk0rixz19FdRGLWdpe7NJPx7997G6i2F1Fzd0fo6C546v5DslDYbSoNP2QjtWjS1S2rW1rTrq7VC3GgqvIuGEywZyCD1hIQTdaPdtCl+sL5WMm3tDABQ12b5eN56b6C6e0zynx/C9LwvBelqJvbsy1WXdbfU72q1vYUfI2eDqmroNWl2fF3uCdCgp23u/rz6GiU0WBoornMkZJJfJHRPlmeRuteA33FWmI/pHcfkjwjMc+0q83zGp6jD0ZjQibCt1w2HbWsDcHEWnWIxyxjlouPiXarqTcLppuyFaYK5lH43+42nffO7zbD3X62o9lUDaeydCby1OvYWds/wButz6inc1+7V1nZGTZglOROVjM5SFlDR7BV3FyLU6rRJTB8Zn0AAAAAAAANCGsaZ3HjWOjfDBjR+9Gu+9S+aN/0a5vMlrexOdVOuT4DaNg2+x3g92yRk2o6zeYgZYzJKMSkk5ZWXRIx/hxHzlJmbfeAAFJXlLoM5L9U+GuYqFKlpRpCeRxK2XuUrdcePW8QwxoTZMRmy257Fs1UmDTBsR0ZmZmVUkfXDJl9Z9cIJCH3kLtUxwv5Yl+6LtqndGwtF738b915aaT+pNfWLaCle3ZUbzJ3epVGTgqjHv5WCzfGUzhpHzUmkSHWdoGKd2gRk8MlYb4Leb9l8m+KLjbR+4oV3WNmV/X0xYbdVpFD6tJ1aQ2FeLTsFGtSzf21MoTEFH2dnGzCGTZyhKNnaPpj3foLaQHTtiXaL1pQLxsWbayz6HoVRsdylWMDGPJucex1Zh3k08ZwsNHIuJCWlnTdkohHRjFBZ2+eKItWySiypC509/Gf29G9ueTSH6g7b032JXegZqVuOoOItHyvOGyY3m3k7WEgwcPJK72TYE8zbwy+5thRUU5ZWW7YRasmTN5iAYYUSdMkYqMfTWodyxOs/JF4fWGjd5yu9+1/K7VOjdB3VnrixT2qpzR2w9w663HZ9kTO0I5kWnw0LryNpc4xnI6SkmsszOkkz+rKrouDJ75jJthmzaM8Hyphq2QbYPnHs5PhBIiWD5xjOcYybBfXOPXPp6+nrkfSAgb5CunZzlPTETfVOd9k9AapsNyjdf75PqR1LOb3qXVFwaPYqe2rG1CuwcvZ7oyrxlUUZONrbiKmI9F4SXbu8JtFcp6zemuS9ydI+Lvzh6C5i1jtuq8a7YuS9m8b9A2hAy9P2hbZKus63sDZSEc2vzNvfXVOveyqq3iKw4viysk7Qduo9B4gdJ0qlKvx7We09x+VjRPW1U1DuTWmsuS/GFDc1bZfbd1jadbJk6GuVmi5CX1vWC2xpGuLGeot4mVVlZeIQeRZE0WecuiYfs8uNqsAGv15bORPINu7eutt+c7E5WvenOdNB7mUg9S9AzW2Wbxrui/V6ZgZvasBXtdVeVYWi51/Xpf4BrM87KJJV+YnJx41YFePEHyVSsBortM3gZ8W2yNnadxNTnBnXvLnUUFqfT0NeHOznvLGs4l3DSymyqNa4tlNr7cg3lwukxP1apNlY8kLF155HmWcklUC2ueHqOsG6+4PLt5E42k7DpGhuvdicuU/QudoUed15arlEc4abkKHdbmWrWZuxnWEC9ssjiPh1H8e1UcnYSGfZ9UclLsJgA0xPL9RuQbPtryIUWf8YW8nPkR3BEaea8ddR0ul33aC+974lS6nAUq8602JBxhq7oJXUMgwia1e4p1LxKsnD1nLly4fYURaJyH6Pkdq+Pvvzj3vPpCg7n2xUp/xKH4n3LcNUUme2y8hekalP52g2LKQlPYyEs2PsickJmLRsTlLMM4dIpGVctitl1T2V+Arm/Z/KPiY5K05uWvyFS2U0h9k3qy1KXamZS9WztjcWwdnw0BMMznOo0mYyv26JbzDRX2VWkmV21VTTOiYmLiQAfzzOWuQ57mqX0ByDQub+m6/wCVHRHlHhLGl0KjXb0eiPeF3Ox1LFcLG+26hlprc+lLZrKQeYsWvEHai0xeJSRkVIr+Kyj9FKZ3kNiNm84bc86/PSmht97Isflj19pRzx1Na61xYb5WLzeJKl41VcaZJ2eBZKQlGVpckdKVdp255GFxEZy8I4URXanX27OWNXy+keZeedNz7pu9ntU6S1brqbds/X6o5mKbSYSvSa7X1znOWyj2PXMhnOc5ylkmc59c5GeQGI9/bCl9S6M3HtKv159bZ/XWr75doOrRjVw9kLHM1isSczFwbNo0RcOnLiVfM27JJFsgsuoZfBUkVFMlIaoHhrje4OvBVLaFsbeXZ7/7C5T37a9yz0ukeIuc5vframXSUsUxZHEoybO0J6OfXSPgkTyzT3kUxg45gol9XYETxUNwGx2h0v0z4INRk0Vu7WVn8Qui9/1rsd/sjWFnqFRqtucakZ6FptXr1sm2rSFuElbJaDLYGi1adSjXMMqu5KpkiCxkt1MAGtV5yapDWPpzx+P+s9Z7Zv8A406ubeEpuZ9o+r7As1pq++XtVbRWpJC7/wC1DFzsqLozhk6lmTBWqLIIvZxwu3sGFWKTZFWudpz11jM+CjmG0uNMbafR/Enks1519obSsjV8R+75PiTTG3bQvVI+201BohLyloY6/tUvYE4zLNCySTKLi3qhHqxsZe2neJJKb6E8hflV8jEBR9kUrn3o9TlzVWlHO1KHYNb2a7n0TruViL1Z2lXtLZhPNoFGalWkYxdvI5um7covSpHOdqsRLYfABqAebWm64rfS3SJahU/JFq3efU/HbCgqSXI9LYW3SHd1gjW07Ude6k2A7haZbbNUbZr1aVSaWadWlqM6kNcPMQzGVXbuU8H+6+c57Q8bMF9H06K2nQtjXqlcG6g2HoDqfGqKzLbLk9bO9z6lhGKNnPUam2lbDMw8LcYheurycMi+as01ECmKbDxmZSx7wA6k2JReWeiNu7Eplo138Ynd/UnWtBpN4hXNcucLqza9rYYoS9ng3ucPImTnIiFxPkYO0kXCEdJsPfJEVMfGL0wAAAAAAAA9Mevr6Y9fT09f/Pp/16/9AAAAZxjP/OMZ/wDn7QAAAPTHr6+mPX/v/wA//wCgAB/z/wAhjGMfZjHpj/rAYxjH/GMY9c+ufT/zn/v/AOQAAAMYxj7MY9Mf9YAAHj0xn0znGM5x/wAZ9P8Aj/4/6HnOMZ/5xjP/AM/aAAAemPX19Mevp6ev/n0/69f+h49MZ9M5xjOcf8Z9P+P/AI/6HkAAB49MY9c4xjGc/wDOfT/n/wCf+x5AAzjGcemcYzj/AKz9uADGMYx6YxjGP+sfZgAAPTGfT1xjPp/x/wDj/wCAzjGfszj1x/1kAAAAAAAAAAAAAAAAAAAAAAB8EhKxcSmkrKyTCMSXWI2RVkHjdmms4U9fdoJHcqJlUWP6Z9hImcnN6Z9kufTI+8AHxvpGPi2+Xcm+ZxzQp00zOnzlBo3KosfCaSeVnB008HVUMUiZcm9o584KXGTZxgfYAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA0sfpGGxNZ9bddUDgi09S6v5qrvLXL+5ev5ib2Ns2ta2aXHp+arhYflXXbGRscpHJLSUa+QcWp83ZpyDzMDYFlCsimy2XFgTfzU3Cc8Q3CfVWkYKjX3qLsXZWjuO63EXeQkEdf1np+2HlK5e5G9ZgPdyv8AgH1NtFhTjmysc8fs3ERj3iBHXsGz3yz3j2e38g29+F+0IPns+NBcb1rpR3svQzW8N2V5NP2tWO+vpQ1xfLu6wgyjmspFvK2fMworJRJJtpOFZS6cPH14cs/SBN57t2fzPsmZb8uT/ADf1r0mfQdf571o4v8h1toap2OfnK1rndOz5GRVLTpqHdykOzzeIWMiWJoNhYI10wXVMRVPFnH0gLZkDqPxpbCu9k1Jrvd8Uy3BzXHq6+2kSzqU5+vN74oUQ1lHhKhY6rOZf19w8Tm4fKMyk2LKMWhnrd40wq1Vj10d5EfIe8668jvM/J9Y5gjofhjQ+p+g/9abojdgyj+chbTrqeuExRkIeqTTFB7Pzr+EekiZ1w5i4yvx0cog6YS7uTRdR9svj26jf9rcTcz9WStZaU6X3lqiuXmZrEe5Wdx0NMvklG0uzjXDgxnKkbiSaOlI/Lk53JWZ0COFFFinOanXcvmS6E0dRPJVrOy0DWsp2XzZ1xqvQ/JlIaR1ibVncdF6hf1onOFmsUSSadzjmWUi1Lm9un8DetGSisCVKOTaFMc2Pt788t2+9IdaxPEmmrzyHqnYmu+YofoLdO0upktgEp1wtM64NH1nUWra9TnzZ2ye2BdhIyb+el5Z+nDR7pkiRus4arndZXunlL3HfPGpyr2Pq57y3yzbt+TTCC2Q+7IuFjYVTWLqMSs8fcGmuqdW0mlu3XZndmrKpadVohzGychVHKlgVRXXaGjlIpa/84nQ834qPIl1pKVXT9i3zwVuiU1IhM1+sbDrOp9sskJCgPYm3IUe5v47YFYLIV26mSVipV21dJPGiDzKKSax2Zc6dH+QXyU8nePCt9U7d1poKUv8At/d2rGzCQ1RTNx7M13ypzbtKrRdgdX/csDDqNbxsacoMlmRrMotT28LESDx/BSJE00l1o4sxvEv3Bce3tWbStFu3RyJvdOj7BJX6tfuVZS9RTiSrDyOy5am27pzY7T/VWn7mR83ft2kWvMzTCfjm5pGPVQTbnM5thAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABV7qDxP81Uze/XfR2569T+p9o9bbShb7KzO8dV6/tZda1epV0tWpesqGSaipo7Ct16IxlAzkqiDqVyRpl8nnDFrhOtLsLxK07nLinq6n6boXRHQVG2T2tB9raw01zlKVLXm3uQb+8dNMTNw5jM5j5SNsjCouWLGQjdZFiIxVaKcTLaJcFeK5WN0jw9cxb4mfIj1N15s+G7JndT23kbVegDbU76rEJQ9zbp2Ojan89bVIPWkMm1ZVjW9SrrWKrTRqhFMI5w4Nh2RxOS8jPyeZ7c1+HF3y/sLWzHX/AHF00y5I0psy37V1JyDCOa1S6nCy9teS7/NKuewakyirpsjUUE8n5Z1D61syikJ9YNHrvzPFWGDrzN8inD1X8iPLtk5fuF6ntcQdkuesLktaa1GR8vLNnOsNg17YDFkkxlFUmZ0JR3XkY52oc+FEGzhRVHBlSFxnoSHjipqHSXfPSONlWc013vonX2irXWsw8ViL1/Fa/pNqpTSerzzC31uVkJFtal3zlrJESbIrtEUkjGTUPnGeuJOWoLibk/RXKNZtcteIDRdEY0WLts6xZxsvPNWLh24I/kGEedRk1cKZdGKZJuoZPGCYzjPrnIoavPIc93D9IM1X00pzfvLVOjuI6C9Z7V2JtKvNaxrHo/emsLVfI/nuU15GfXpFvfm9LZ32YusFfDN03kdHJxbDJoZ0hhk6ss688X7rfnRTPrTQfUWz+O+gpHTr3n/Y911xVaLcWmyNTryh5mLj5WHucY4LF22qyK7paqXSKeIycQVx7tNI+UGyiXCdEeJmP3JrXh2Dp/VO96JujgSyK2jT/RV2cQ++r7ZpWYqmKlcJDarLZ6MhF3qVszIiTrMk+wkeJdFOSPSTZqnbDEyPg1oOOTvILynKdM7gtcf5CtrJbk2Hs23QdOkb5Wbk6Z00licsf4W0g4KXbzUjT0nyLZWKjEIhF8ePapqItklDTu6P4xlt4c96w0xRek95c523TbugS9F2vp2xqQz57Ka+h04ZlHbIpSi2altGgTrcplLHrq3tX1dkl8NlzJlWZoKYxxwD43a5xBbuk9wS2157d+/OtLJQ7BufY0hSaJq6Bepawrbqp0eJq2tNbRMTVK02YRL56tLu0U3T+wTDx1LP3Hv1jFFlgAAAAAAAAAAAAAAD/9k="},"smvage.gif":{"image/gif":"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"}},"execution_count":null},{"metadata":{"_kg_hide-output":true,"trusted":true},"cell_type":"code","source":"!pip install tensorflow_io","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"import pandas as pd \nimport os\nimport matplotlib.pyplot as plt \nimport numpy as np \nimport tensorflow as tf\nimport tensorflow_io as tfio \nimport matplotlib.image as mpimg\nimport seaborn as sns \nimport pydicom\nsns.set_palette(\"bright\")","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# Demographic analysis\n ","execution_count":null},{"metadata":{"trusted":true},"cell_type":"code","source":"df=pd.read_csv(\"../input/osic-pulmonary-fibrosis-progression/train.csv\")","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"(sns.countplot(df.Sex)).figure.savefig(\"output1.png\")\n","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"fig, axs = plt.subplots(ncols=3)\nfig.set_size_inches(19,6)\nsns.countplot(df['SmokingStatus'],ax=axs[0])\nsns.countplot(df['SmokingStatus'][df['Sex']==\"Male\"],ax=axs[1])\nsns.countplot(df['SmokingStatus'][df['Sex']==\"Female\"],ax=axs[2])\nfig.savefig(\"output2.jpeg\")","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"fig, ax = plt.subplots()\nfig.set_size_inches(11.7, 8.27)\nsns.distplot(df.Age,kde=False,bins=80,color=\"k\")\nfig.savefig(\"output3.jpeg\")","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"def create_last_scan_df():\n    last_test=pd.DataFrame()    \n    for i in df.Patient.unique():\n        last_test=last_test.append((df[df['Patient']==\"{}\".format(i)][-1:]))\n    last_test=last_test.drop(\"Patient\",axis=1)\n    last_test=last_test.drop(\"Weeks\",axis=1)\n    return last_test\ndd=create_last_scan_df()\ndd=dd.reset_index(drop=True)\ndd.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"def create_baseline():\n    first_scan=pd.DataFrame()    \n    for i in df.Patient.unique():\n        first_scan=first_scan.append((df[df['Patient']==\"{}\".format(i)][:1]))\n    first_scan=first_scan.drop(\"Patient\",axis=1)\n    first_scan=first_scan.drop(\"Weeks\",axis=1)\n    return first_scan\nfc=create_baseline()\nfc=fc.reset_index(drop=True)\nfc.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"(sns.pairplot(df,hue=\"SmokingStatus\",height=4)).savefig(\"output4.jpeg\")\n","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Baseline Data ","execution_count":null},{"metadata":{"trusted":true},"cell_type":"code","source":"sns.pairplot(fc,hue=\"SmokingStatus\",height=4).savefig(\"output5.jpeg\")","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"(sns.pairplot(dd,hue=\"Sex\",height=4)).savefig(\"output6.jpeg\")","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Last Recorded Values","execution_count":null},{"metadata":{"trusted":true},"cell_type":"code","source":"(sns.pairplot(dd,hue=\"SmokingStatus\",height=4)).savefig(\"output7.jpeg\")\n","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"(sns.pairplot(fc,hue=\"Sex\",height=4)).savefig(\"output8.jpeg\")\n","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"ex_smoker_male=df.loc[df['Patient']=='{}'.format((df.loc[(df[\"Sex\"]==\"Male\") & (df[\"SmokingStatus\"]==\"Ex-smoker\"),'Patient'][:1].values[0]))]\nex_smoker_female=df.loc[df['Patient']=='{}'.format((df.loc[(df[\"Sex\"]==\"Female\") & (df[\"SmokingStatus\"]==\"Ex-smoker\"),'Patient'][:1].values[0]))]\nnon_smoker_male=df.loc[df['Patient']=='{}'.format((df.loc[(df[\"Sex\"]==\"Male\") & (df[\"SmokingStatus\"]==\"Never smoked\"),'Patient'][:1].values[0]))]\nnon_smoker_female=df.loc[df['Patient']=='{}'.format((df.loc[(df[\"Sex\"]==\"Female\") & (df[\"SmokingStatus\"]==\"Never smoked\"),'Patient'][:1].values[0]))]\ncurrent_smoker_male=df.loc[df['Patient']=='{}'.format((df.loc[(df[\"Sex\"]==\"Male\") & (df[\"SmokingStatus\"]==\"Currently smokes\"),'Patient'][:1].values[0]))]\ncurrent_smoker_female=df.loc[df['Patient']=='{}'.format((df.loc[(df[\"Sex\"]==\"Female\") & (df[\"SmokingStatus\"]==\"Currently smokes\"),'Patient'][:1].values[0]))]","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# FVC and Percent Trend For All Patients","execution_count":null},{"metadata":{"trusted":true},"cell_type":"code","source":"fig, ax = plt.subplots(nrows=2)\nfig.set_size_inches(22, 8.27)\nsns.lineplot(x='Weeks',y='Percent',data=df,ax=ax[0])\nsns.lineplot(x='Weeks',y='FVC',data=df,ax=ax[1])\nfig.savefig(\"weeksvsfvc.jpeg\")\n","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# FVC And Percent Trend Male Vs Female","execution_count":null},{"metadata":{"trusted":true},"cell_type":"code","source":"males=df[df[\"Sex\"]==\"Male\"]\nfemales=df[df[\"Sex\"]==\"Female\"]","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"fig, ax = plt.subplots(nrows=4)\nfig.set_size_inches(22, 22)\nsns.lineplot(x='Weeks',y='FVC',data=males,ax=ax[0]).set_title(\"MALES FVC TREND\")\nsns.lineplot(x='Weeks',y='FVC',data=females,ax=ax[1]).set_title(\"FEMALE FVC TREND\")\nsns.lineplot(x='Weeks',y='Percent',data=males,ax=ax[2]).set_title(\"MALES PERCENT TREND\")\nsns.lineplot(x='Weeks',y='Percent',data=females,ax=ax[3]).set_title(\"FEMALE PERCENT TREND\")\nfig.savefig(\"mvffvctrend.jpeg\")\n","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# FVC And Percent Tread Of All Patients Smoker Vs Non-Smoker\n","execution_count":null},{"metadata":{"trusted":true},"cell_type":"code","source":"smoker=df[df[\"SmokingStatus\"]==\"Ex-smoker\"]\nnever_smoked=df[df[\"SmokingStatus\"]==\"Never smoked\"]\ncurrent_smoker=df[df[\"SmokingStatus\"]==\"Currently smokes\"]","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"fig, ax = plt.subplots(nrows=6)\nfig.set_size_inches(22, 35)\nsns.lineplot(x='Weeks',y='FVC',data=smoker,ax=ax[0]).set_title(\"EX SMOKER FVC TREND\")\nsns.lineplot(x='Weeks',y='FVC',data=never_smoked,ax=ax[1]).set_title(\"NON SMOKER FVC TREND\")\nsns.lineplot(x='Weeks',y='FVC',data=current_smoker,ax=ax[2]).set_title(\"SMOKER FVC TREND\")\nsns.lineplot(x='Weeks',y='Percent',data=smoker,ax=ax[3]).set_title(\"EX SMOKER PERCENT  TREND\")\nsns.lineplot(x='Weeks',y='Percent',data=never_smoked,ax=ax[4]).set_title(\"NON SMOKER PERCENT TREND\")\nsns.lineplot(x='Weeks',y='Percent',data=current_smoker,ax=ax[5]).set_title(\"SMOKER PERCENT TREND\")\nfig.savefig(\"weeksvpercent.jpeg\")\n","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# FVC And Percent Trend Of Random Patients Of All Categories","execution_count":null},{"metadata":{"trusted":true},"cell_type":"code","source":"fig, ax = plt.subplots()\nfig.set_size_inches(22,5)\nsns.lineplot(x=ex_smoker_male[\"Weeks\"], y=ex_smoker_male[\"FVC\"],label='ex_smoker_male')\nsns.lineplot(x=ex_smoker_female[\"Weeks\"], y=ex_smoker_female[\"FVC\"],label='ex_smoker_female')\nsns.lineplot(x=non_smoker_male[\"Weeks\"], y=non_smoker_male[\"FVC\"],label='non_smoker_male')\nsns.lineplot(x=non_smoker_female[\"Weeks\"], y=non_smoker_female[\"FVC\"],label='non_smoker_female')\nsns.lineplot(x=current_smoker_male[\"Weeks\"], y=current_smoker_male[\"FVC\"],label='current_smoker_male')\nsns.lineplot(x=current_smoker_female[\"Weeks\"], y=current_smoker_female[\"FVC\"],label='current_smoker_female')\nfig.savefig(\"smoker_current_fvc.jpeg\")\n","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"fig, ax = plt.subplots()\nfig.set_size_inches(22,5)\nsns.lineplot(x=ex_smoker_male[\"Weeks\"], y=ex_smoker_male[\"Percent\"],label='ex_smoker_male')\nsns.lineplot(x=ex_smoker_female[\"Weeks\"], y=ex_smoker_female[\"Percent\"],label='ex_smoker_female')\nsns.lineplot(x=non_smoker_male[\"Weeks\"], y=non_smoker_male[\"Percent\"],label='non_smoker_male')\nsns.lineplot(x=non_smoker_female[\"Weeks\"], y=non_smoker_female[\"Percent\"],label='non_smoker_female')\nsns.lineplot(x=current_smoker_male[\"Weeks\"], y=current_smoker_male[\"Percent\"],label='current_smoker_male')\nsns.lineplot(x=current_smoker_female[\"Weeks\"], y=current_smoker_female[\"Percent\"],label='current_smoker_female')\nfig.savefig(\"sdad.jpeg\")\n","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# Visualising Dicom Files ","execution_count":null},{"metadata":{"trusted":true},"cell_type":"code","source":"files=[]\nfor dirname, _, filenames in os.walk('../input/osic-pulmonary-fibrosis-progression/train'):\n    for filename in filenames:\n        files.append(os.path.join(dirname, filename))","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"### Using tensorflow_io to decode Dicom files","execution_count":null},{"metadata":{"trusted":true},"cell_type":"code","source":"def decode_image(image_path):\n    image_bytes = tf.io.read_file(image_path)\n    image = tfio.image.decode_dicom_image(image_bytes, dtype=tf.uint16)\n    image=np.squeeze(image.numpy())\n    return image ","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"def show_scan(image):\n    img = decode_image(image)\n    patient_name=str(image).split('/')[1]\n    fig, ax = plt.subplots()\n    im=ax.imshow(img,cmap='twilight_r')\n    plt.axis('off')\n    plt.title(\"Baseline CT Scan of Patient {}\".format(patient_name))\n    fig.set_size_inches(9,9)\n    plt.show()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"show_scan(files[3])","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"def choose_patient(ID):\n    images_x=[]\n    for i in files:\n        name=str(i)\n        if str(ID) in name:\n            images_x.append(i)\n    return sorted(images_x)\n\ndef generate_images(images):\n    for x,i in enumerate(images):\n        image=decode_image(i)   \n        fname=str(x)+\".png\"\n        plt.imsave(fname,image,cmap='twilight_shifted')    \n        \ndef make_progressive_video():\n    os.system(\"ffmpeg  -r 30 -i %d.png -vcodec mpeg4 -y -vb 400M patient_ct_scan_progression.mp4\")","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"patient_x=choose_patient(df['Patient'][288])\ngenerate_images(patient_x)\nmake_progressive_video()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"!rm -r *.png","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_kg_hide-output":true},"cell_type":"code","source":"!ffmpeg -i patient_ct_scan_progression.mp4 out.gif","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"![Gif](./out.gif)","execution_count":null}],"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"pygments_lexer":"ipython3","nbconvert_exporter":"python","version":"3.6.4","file_extension":".py","codemirror_mode":{"name":"ipython","version":3},"name":"python","mimetype":"text/x-python"}},"nbformat":4,"nbformat_minor":4}