📌 Data is collected for the Omdena's Local Chapter project- Creating a Rice Disease Classifier using Open Source Data and Computer Vision

All the images are stored in 224 x 224 pixels form for the ease of training the models.

1. BACKGROUND - Rice is the heart of Filipino cuisine, an ingredient so fundamental to the country’s food culture that it is eaten at every meal, from breakfast to dinner. - It’s served at every meal with almost every dish, including meats, seafood, vegetables, and soups. Rice also plays a significant role in Filipino celebrations and rituals and is a symbol of community and shared identity.

2. PROBLEM - A variety of fungal, bacterial, and viral diseases can infect rice plants, causing reduced yield, lower quality, and even total crop loss. These diseases can be devastating to farmers, especially those with limited resources who cannot afford the cost of chemical treatments or disease-resistant seeds. - Additionally, the high humidity and frequent rain in the Philippines create favorable conditions for the growth and spread of rice diseases.

3. RESEARCH We came across 13 rice diseases that are spread into three categories: - Fungal- affects the leaf blade, sheath, stem, node, and panicle - Bacterial- affects the leaf blade - Viral- affects the leaf blade and sheath

⚡ Click here to know the parts of a Rice Plant

List of all 13 diseases under three categories: - i. Fungal-- Rice Blast, Sheath Blight, Brown Spot, Narrow Brown Spot, Sheath Rot, Stem Rot, Bakanae, & Rice False Smut - ii. Bacterial-- Bacterial Leaf Blight & Bacterial Leaf Streak - iii. Viral-- Tungro Virus, Ragged Stunt Virus, & Grassy Stunt Virus

4. SOLUTION To build a deep-learning classifier model that can accurately detect different rice diseases.

Project Tycho datasets contain case counts for reported disease conditions for countries around the world. The Project Tycho data curation team extracts these case counts from various reputable sources, typically from national or international health authorities, such as the US Centers for Disease Control or the World Health Organization. These original data sources include both open- and restricted-access sources. For restricted-access sources, the Project Tycho team has obtained permission for redistribution from data contributors. All datasets contain case count data that are identical to counts published in the original source and no counts have been modified in any way by the Project Tycho team. The Project Tycho team has pre-processed datasets by adding new variables, such as standard disease and location identifiers, that improve data interpretabilty. We also formatted the data into a standard data format.

Each Project Tycho dataset contains case counts for a specific condition (e.g. measles) and for a specific country (e.g. The United States). Case counts are reported per time interval. In addition to case counts, datsets include information about these counts (attributes), such as the location, age group, subpopulation, diagnostic certainty, place of aquisition, and the source from which we extracted case counts. One dataset can include many series of case count time intervals, such as "US measles cases as reported by CDC", or "US measles cases reported by WHO", or "US measles cases that originated abroad", etc.

Depending on the intended use of a dataset, we recommend a few data processing steps before analysis:

• Analyze missing data: Project Tycho datasets do not inlcude time intervals for which no case count was reported (for many datasets, time series of case counts are incomplete, due to incompleteness of source documents) and users will need to add time intervals for which no count value is available. Project Tycho datasets do include time intervals for which a case count value of zero was reported.
• Separate cumulative from non-cumulative time interval series. Case count time series in Project Tycho datasets can be "cumulative" or "fixed-intervals". Cumulative case count time series consist of overlapping case count intervals starting on the same date, but ending on different dates. For example, each interval in a cumulative count time series can start on January 1st, but end on January 7th, 14th, 21st, etc. It is common practice among public health agencies to report cases for cumulative time intervals. Case count series with fixed time intervals consist of mutually exxclusive time intervals that all start and end on different dates and all have identical length (day, week, month, year). Given the different nature of these two types of case count data, we indicated this with an attribute for each count value, named "PartOfCumulativeCountSeries".

Project Tycho datasets contain case counts for reported disease conditions for countries around the world. The Project Tycho data curation team extracts these case counts from various reputable sources, typically from national or international health authorities, such as the US Centers for Disease Control or the World Health Organization. These original data sources include both open- and restricted-access sources. For restricted-access sources, the Project Tycho team has obtained permission for redistribution from data contributors. All datasets contain case count data that are identical to counts published in the original source and no counts have been modified in any way by the Project Tycho team. The Project Tycho team has pre-processed datasets by adding new variables, such as standard disease and location identifiers, that improve data interpretabilty. We also formatted the data into a standard data format.

Each Project Tycho dataset contains case counts for a specific condition (e.g. measles) and for a specific country (e.g. The United States). Case counts are reported per time interval. In addition to case counts, datsets include information about these counts (attributes), such as the location, age group, subpopulation, diagnostic certainty, place of aquisition, and the source from which we extracted case counts. One dataset can include many series of case count time intervals, such as "US measles cases as reported by CDC", or "US measles cases reported by WHO", or "US measles cases that originated abroad", etc.

Depending on the intended use of a dataset, we recommend a few data processing steps before analysis:

• Analyze missing data: Project Tycho datasets do not inlcude time intervals for which no case count was reported (for many datasets, time series of case counts are incomplete, due to incompleteness of source documents) and users will need to add time intervals for which no count value is available. Project Tycho datasets do include time intervals for which a case count value of zero was reported.
• Separate cumulative from non-cumulative time interval series. Case count time series in Project Tycho datasets can be "cumulative" or "fixed-intervals". Cumulative case count time series consist of overlapping case count intervals starting on the same date, but ending on different dates. For example, each interval in a cumulative count time series can start on January 1st, but end on January 7th, 14th, 21st, etc. It is common practice among public health agencies to report cases for cumulative time intervals. Case count series with fixed time intervals consist of mutually exxclusive time intervals that all start and end on different dates and all have identical length (day, week, month, year). Given the different nature of these two types of case count data, we indicated this with an attribute for each count value, named "PartOfCumulativeCountSeries".

The 2008 National Demographic and Health Survey (2008 NDHS) is a nationally representative survey of 13,594 women age 15-49 from 12,469 households successfully interviewed, covering 794 enumeration areas (clusters) throughout the Philippines. This survey is the ninth in a series of demographic and health surveys conducted to assess the demographic and health situation in the country. The survey obtained detailed information on fertility levels, marriage, fertility preferences, awareness and use of family planning methods, breastfeeding practices, nutritional status of women and young children, childhood mortality, maternal and child health, and knowledge and attitudes regarding HIV/AIDS and tuberculosis. Also, for the first time, the Philippines NDHS gathered information on violence against women. The 2008 NDHS was conducted by the Philippine National Statistics Office (NSO). Technical assistance was provided by ICF Macro through the MEASURE DHS program. Funding for the survey was mainly provided by the Government of the Philippines. Financial support for some preparatory and processing phases of the survey was provided by the U.S. Agency for International Development (USAID). Like previous Demographic and Health Surveys (DHS) conducted in the Philippines, the 2008 National Demographic and Health Survey (NDHS) was primarily designed to provide information on population, family planning, and health to be used in evaluating and designing policies, programs, and strategies for improving health and family planning services in the country. The 2008 NDHS also included questions on domestic violence. Specifically, the 2008 NDHS had the following objectives: Collect data at the national level that will allow the estimation of demographic rates, particularly, fertility rates by urban-rural residence and region, and under-five mortality rates at the national level. Analyze the direct and indirect factors which determine the levels and patterns of fertility. Measure the level of contraceptive knowledge and practice by method, urban-rural residence, and region. Collect data on family health: immunizations, prenatal and postnatal checkups, assistance at delivery, breastfeeding, and prevalence and treatment of diarrhea, fever, and acute respiratory infections among children under five years. Collect data on environmental health, utilization of health facilities, prevalence of common noncommunicable and infectious diseases, and membership in health insurance plans. Collect data on awareness of tuberculosis. Determine women's knowledge about HIV/AIDS and access to HIV testing. Determine the extent of violence against women. MAIN RESULTS FERTILITY Fertility Levels and Trends. There has been a steady decline in fertility in the Philippines in the past 36 years. From 6.0 children per woman in 1970, the total fertility rate (TFR) in the Philippines declined to 3.3 children per woman in 2006. The current fertility level in the country is relatively high compared with other countries in Southeast Asia, such as Thailand, Singapore and Indonesia, where the TFR is below 2 children per woman. Fertility Differentials. Fertility varies substantially across subgroups of women. Urban women have, on average, 2.8 children compared with 3.8 children per woman in rural areas. The level of fertility has a negative relationship with education; the fertility rate of women who have attended college (2.3 children per woman) is about half that of women who have been to elementary school (4.5 children per woman). Fertility also decreases with household wealth: women in wealthier households have fewer children than those in poorer households. FAMILY PLANNING Knowledge of Contraception. Knowledge of family planning is universal in the Philippines- almost all women know at least one method of fam-ily planning. At least 90 percent of currently married women have heard of the pill, male condoms, injectables, and female sterilization, while 87 percent know about the IUD and 68 percent know about male sterilization. On average, currently married women know eight methods of family planning. Unmet Need for Family Planning. Unmet need for family planning is defined as the percentage of currently married women who either do not want any more children or want to wait before having their next birth, but are not using any method of family planning. The 2008 NDHS data show that the total unmet need for family planning in the Philippines is 22 percent, of which 13 percent is limiting and 9 percent is for spacing. The level of unmet need has increased from 17 percent in 2003. Overall, the total demand for family planning in the Philippines is 73 percent, of which 69 percent has been satisfied. If all of need were satisfied, a contraceptive prevalence rate of about 73 percent could, theoretically, be expected. Comparison with the 2003 NDHS indicates that the percentage of demand satisfied has declined from 75 percent. MATERNAL HEALTH Antenatal Care. Nine in ten Filipino mothers received some antenatal care (ANC) from a medical professional, either a nurse or midwife (52 percent) or a doctor (39 percent). Most women have at least four antenatal care visits. More than half (54 percent) of women had an antenatal care visit during the first trimester of pregnancy, as recommended. While more than 90 percent of women who received antenatal care had their blood pressure monitored and weight measured, only 54 percent had their urine sample taken and 47 percent had their blood sample taken. About seven in ten women were informed of pregnancy complications. Three in four births in the Philippines are protected against neonatal tetanus. Delivery and Postnatal Care. Only 44 percent of births in the Philippines occur in health facilities-27 percent in a public facility and 18 percent in a private facility. More than half (56 percent) of births are still delivered at home. Sixty-two percent of births are assisted by a health professional-35 percent by a doctor and 27 percent by a midwife or nurse. Thirty-six percent are assisted by a traditional birth attendant or hilot. About 10 percent of births are delivered by C-section. The Department of Health (DOH) recommends that mothers receive a postpartum check within 48 hours of delivery. A majority of women (77 percent) had a postnatal checkup within two days of delivery; 14 percent had a postnatal checkup 3 to 41 days after delivery. CHILD HEALTH Childhood Mortality. Childhood mortality continues to decline in the Philippines. Currently, about one in every 30 children in the Philippines dies before his or her fifth birthday. The infant mortality rate for the five years before the survey (roughly 2004-2008) is 25 deaths per 1,000 live births and the under-five mortality rate is 34 deaths per 1,000 live births. This is lower than the rates of 29 and 40 reported in 2003, respectively. The neonatal mortality rate, representing death in the first month of life, is 16 deaths per 1,000 live births. Under-five mortality decreases as household wealth increases; children from the poorest families are three times more likely to die before the age of five as those from the wealthiest families. There is a strong association between under-five mortality and mother's education. It ranges from 47 deaths per 1,000 live births among children of women with elementary education to 18 deaths per 1,000 live births among children of women who attended college. As in the 2003 NDHS, the highest level of under-five mortality is observed in ARMM (94 deaths per 1,000 live births), while the lowest is observed in NCR (24 deaths per 1,000 live births). NUTRITION Breastfeeding Practices. Eighty-eight percent of children born in the Philippines are breastfed. There has been no change in this practice since 1993. In addition, the median durations of any breastfeeding and of exclusive breastfeeding have remained at 14 months and less than one month, respectively. Although it is recommended that infants should not be given anything other than breast milk until six months of age, only one-third of Filipino children under six months are exclusively breastfed. Complementary foods should be introduced when a child is six months old to reduce the risk of malnutrition. More than half of children ages 6-9 months are eating complementary foods in addition to being breastfed. The Infant and Young Child Feeding (IYCF) guidelines contain specific recommendations for the number of times that young children in various age groups should be fed each day as well as the number of food groups from which they should be fed. NDHS data indicate that just over half of children age 6-23 months (55 percent) were fed according to the IYCF guidelines. HIV/AIDS Awareness of HIV/AIDS. While over 94 percent of women have heard of AIDS, only 53 percent know the two major methods for preventing transmission of HIV (using condoms and limiting sex to one uninfected partner). Only 45 percent of young women age 15-49 know these two methods for preventing HIV transmission. Knowledge of prevention methods is higher in urban areas than in rural areas and increases dramatically with education and wealth. For example, only 16 percent of women with no education know that using condoms limits the risk of HIV infection compared with 69 percent of those who have attended college. TUBERCULOSIS Knowledge of TB. While awareness of tuberculosis (TB) is high, knowledge of its causes and symptoms is less common. Only 1 in 4 women know that TB is caused by microbes, germs or bacteria. Instead, respondents tend to say that TB is caused by smoking or drinking alcohol, or that it is inherited. Symptoms associated with TB are better recognized. Over half of the respondents cited coughing, while 39 percent mentioned weight loss, 35 percent mentioned blood in sputum, and 30 percent cited coughing with sputum. WOMEN'S STATUS Women's Status and Employment.

Philippines PH: Prevalence of Stunting: Height for Age: % of Children Under 5 data was reported at 33.400 % in 2015. This records an increase from the previous number of 30.300 % for 2013. Philippines PH: Prevalence of Stunting: Height for Age: % of Children Under 5 data is updated yearly, averaging 36.050 % from Dec 1987 (Median) to 2015, with 10 observations. The data reached an all-time high of 44.700 % in 1987 and a record low of 30.300 % in 2013. Philippines PH: Prevalence of Stunting: Height for Age: % of Children Under 5 data remains active status in CEIC and is reported by World Bank. The data is categorized under Global Database’s Philippines – Table PH.World Bank.WDI: Health Statistics. Prevalence of stunting is the percentage of children under age 5 whose height for age is more than two standard deviations below the median for the international reference population ages 0-59 months. For children up to two years old height is measured by recumbent length. For older children height is measured by stature while standing. The data are based on the WHO's new child growth standards released in 2006.; ; UNICEF, WHO, World Bank: Joint child malnutrition estimates (JME). Aggregation is based on UNICEF, WHO, and the World Bank harmonized dataset (adjusted, comparable data) and methodology.; Linear mixed-effect model estimates; Undernourished children have lower resistance to infection and are more likely to die from common childhood ailments such as diarrheal diseases and respiratory infections. Frequent illness saps the nutritional status of those who survive, locking them into a vicious cycle of recurring sickness and faltering growth (UNICEF, www.childinfo.org). Estimates of child malnutrition, based on prevalence of underweight and stunting, are from national survey data. The proportion of underweight children is the most common malnutrition indicator. Being even mildly underweight increases the risk of death and inhibits cognitive development in children. And it perpetuates the problem across generations, as malnourished women are more likely to have low-birth-weight babies. Stunting, or being below median height for age, is often used as a proxy for multifaceted deprivation and as an indicator of long-term changes in malnutrition.

Philippines PH: Prevalence of Severe Wasting: Weight for Height: % of Children under 5 data was reported at 2.400 % in 2011. This records an increase from the previous number of 2.000 % for 2008. Philippines PH: Prevalence of Severe Wasting: Weight for Height: % of Children under 5 data is updated yearly, averaging 2.000 % from Dec 2003 (Median) to 2011, with 3 observations. The data reached an all-time high of 2.400 % in 2011 and a record low of 1.600 % in 2003. Philippines PH: Prevalence of Severe Wasting: Weight for Height: % of Children under 5 data remains active status in CEIC and is reported by World Bank. The data is categorized under Global Database’s Philippines – Table PH.World Bank: Health Statistics. Prevalence of severe wasting is the proportion of children under age 5 whose weight for height is more than three standard deviations below the median for the international reference population ages 0-59.; ; UNICEF, WHO, World Bank: Joint child malnutrition estimates (JME). Aggregation is based on UNICEF, WHO, and the World Bank harmonized dataset (adjusted, comparable data) and methodology.; Linear mixed-effect model estimates; Undernourished children have lower resistance to infection and are more likely to die from common childhood ailments such as diarrheal diseases and respiratory infections. Frequent illness saps the nutritional status of those who survive, locking them into a vicious cycle of recurring sickness and faltering growth (UNICEF, www.childinfo.org). Estimates of child malnutrition, based on prevalence of underweight and stunting, are from national survey data. The proportion of underweight children is the most common malnutrition indicator. Being even mildly underweight increases the risk of death and inhibits cognitive development in children. And it perpetuates the problem across generations, as malnourished women are more likely to have low-birth-weight babies. Stunting, or being below median height for age, is often used as a proxy for multifaceted deprivation and as an indicator of long-term changes in malnutrition.

Philippines PH: Prevalence of Wasting: Weight for Height: % of Children Under 5 data was reported at 7.100 % in 2015. This records a decrease from the previous number of 7.900 % for 2013. Philippines PH: Prevalence of Wasting: Weight for Height: % of Children Under 5 data is updated yearly, averaging 7.200 % from Dec 1987 (Median) to 2015, with 10 observations. The data reached an all-time high of 9.100 % in 1993 and a record low of 5.700 % in 1987. Philippines PH: Prevalence of Wasting: Weight for Height: % of Children Under 5 data remains active status in CEIC and is reported by World Bank. The data is categorized under Global Database’s Philippines – Table PH.World Bank: Health Statistics. Prevalence of wasting is the proportion of children under age 5 whose weight for height is more than two standard deviations below the median for the international reference population ages 0-59.; ; UNICEF, WHO, World Bank: Joint child malnutrition estimates (JME). Aggregation is based on UNICEF, WHO, and the World Bank harmonized dataset (adjusted, comparable data) and methodology.; Linear mixed-effect model estimates; Undernourished children have lower resistance to infection and are more likely to die from common childhood ailments such as diarrheal diseases and respiratory infections. Frequent illness saps the nutritional status of those who survive, locking them into a vicious cycle of recurring sickness and faltering growth (UNICEF, www.childinfo.org). Estimates of child malnutrition, based on prevalence of underweight and stunting, are from national survey data. The proportion of underweight children is the most common malnutrition indicator. Being even mildly underweight increases the risk of death and inhibits cognitive development in children. And it perpetuates the problem across generations, as malnourished women are more likely to have low-birth-weight babies. Stunting, or being below median height for age, is often used as a proxy for multifaceted deprivation and as an indicator of long-term changes in malnutrition.

Background: Inflammatory bowel disease (IBD), though historically common in the West, is now increasingly prevalent in industrializing countries. A simultaneous dramatic increase in IBD drug options has enabled most patients to achieve remission. Nevertheless, worldwide disparities in the approval of IBD drugs, or “drug lag”, remain problematic. Summary: Drug lag for major IBD drugs before March 31, 2018 (12 for Crohn’s disease [CD] and 13 for ulcerative colitis [UC]) was compared between that of the United States (US), European Union (EU), and Asia (Japan, China, South Korea, Taiwan, and the Philippines) to assess current trends. In the US, unapproved IBD drugs accounted for 16.7% (2/12) for CD and 23.1% (3/13) for UC; approval lag was 3.8 (0–80.5) months for CD and 3.6 (0–88) months for UC. In the EU, unapproved drugs accounted for 16.7% (2/12) for CD and 15.4% (2/13) for UC; approval lag was 0.03 (0–13.9) months for CD and 0 (0–13.9) months for UC. This demonstrates the short drug lag in both regions, although one drug developed in a joint US/EU clinical trial had around a 350-day approval lag. In Asia, the proportion of unapproved IBD drugs was the lowest in Japan at 33.3% (4/12) for CD and 23.1% (3/13) for UC; South Korea had the shortest lag for CD at 13.2 (0–133.1) months and the Philippines had the shortest lag for UC at 9.9 (0.6–176.2) months, but these countries still had longer lag periods than the West. However, a proportion of unapproved drugs and approval lag has decreased considerably in Asia since the start of the biologics era. Key Messages: Despite the recent shortening drug lag between different countries and regions, this study shows that disparities persist. With globalization, eliminating these disparate drug lags is necessary to manage IBD and may require efforts toward international adoption of a more standardized approval process.