AI and dairy

Why did I write the other day about an artificial intelligence dude who I knew 25 years ago, and whose primary application at the time was ensuring elevators in skyscrapers were efficiently dispersed to floors that needed them – oh, and vision?

Because he made the N.Y Times with an hyperbaric headline about making Toronto a high-tech hotbed (he didn’t write the headline) and because his AI basics are making their way into food safety.

Caroline Diana of Inquisitr writes IBM and Cornell University, which primarily focuses on dairy research, will make use of artificial intelligence (AI) to make dairy safe(r) for consumption.

By sequencing and analyzing the DNA and RNA of food microbiomes, researchers plan to create new tools that can help monitor raw milk to detect anomalies that represent food safety hazards and possible fraud.

While many food producers already have rigorous processes in place to ensure food safety hazards are managed appropriately, this pioneering application of genomics will be designed to enable a deeper understanding and characterization of microorganisms on a much larger scale than has previously been possible.

Only a PR thingy could have written this paragraph: “This work could eventually be extended to the larger context of the food supply chain — from farm to fork — and, using artificial intelligence and machine learning, may lead to new insights into how microorganisms interact within a particular environment. A carefully designed informatics infrastructure developed in the IBM Accelerated Discovery Lab, a data and analytics hub for IBM researchers and their clients and partners, will help the team parse and aggregate terabytes of genomic data.”

Better than a poorly designed informatics infrastructure.

E. coli O26, HUS and dairy

In their recent article in Eurosurveillance, Germinario et al. describe a community-wide outbreak of Shiga toxin 2-producing Escherichia coli (STEC) O26:H11 infections associated with haemolytic uraemic syndrome (HUS) and involving 20 children between 11 and 78 months of age in southern Italy during the summer 2013 [1]. The investigation identified an association between STEC infection and consumption of dairy products from two local milk-processing establishments. We underline striking similarities to a recent multi-country STEC O26 outbreak in Romania and Italy and discuss the challenges that STEC infections and their surveillance pose at the European level.

e-coli-colbertIn March 2016, Peron et al. published, also in Eurosurveillance, early findings of the investigation of a community-wide STEC infection outbreak in southern Romania [2]. As at 29 February 2016, 15 HUS cases with onset of symptoms after 24 January 2016, all but one in children less than two years of age, had been identified, three of whom had died. Aetiological confirmation was retrospectively performed through serological diagnosis and six cases were confirmed with STEC O26 infection. Shortly after this publication, and following the identification of the first epidemiologically-linked case in central Italy, the European Centre for Disease Prevention and Control (ECDC) and the European Food Safety Authority (EFSA) published a joint Rapid Outbreak Assessment [3]. The Italian and Romanian epidemiological, microbiological and environmental investigations implicated products from a milk-processing establishment in southern Romania as a possible source of infection. The dairy plant exported milk products to at least four European Union (EU) countries. The plant was closed in March 2016 and the implicated food products recalled or withdrawn from the retail market.

Pulsed Field Gel Electrophoresis (PFGE) and whole genome sequencing (WGS) analyses did not establish a microbiological link between the Italian (2013) and the Romanian/Italian (2016) outbreaks (personal communication, Stefano Morabito, October 2016). However, the epidemiological similarities between the two community-wide outbreaks associated with HUS and STEC O26 infections, mostly affecting young children and implicating dairy products, are notable. While raw milk and unpasteurised dairy products are well known potential sources of STEC infection, milk products, as highlighted by Germinaro et al. [1], have been rarely implicated in community-wide STEC outbreaks in the past, emphasising an emerging risk of STEC O26 infection associated with milk products.

Reporting of STEC O26 infections has been steadily increasing in the EU since 2007, partly due to improved diagnostics of non-O157 sero-pathotypes [4]. The attention to non-O157 STEC sero-pathotypes rose considerably after the severe STEC O104 outbreak that took place in Germany and France in 2011 during which almost 4,000 cases and more than 50 deaths were reported [5]. In light of the recently published outbreaks related to dairy products and the simultaneous increased reporting of isolations of STEC O26 from milk and milk products in the EU/European Economic Area (EEA) [6], strengthening STEC surveillance in humans and food and enhancing HUS surveillance in children less than five years of age is warranted. Paediatric nephrologists should be sensitised to this effect

Community-wide outbreaks of haemolytic uraemic syndrome associated with Shiga-toxin producing Escherichia coli O26 in Italy and Romania: A new challenge for the European Union

Eurosurveillance, Volume 21, Issue 49, 08 December 2016, DOI:

E Severi, F Vial, E Peron, O Mardh, T Niskanen, J Takkinen

Over 1000 sickened with links to South African dairy

Malibongwe Dayimani of Dispatch Live reports a King William’s Town dairy that supplied a number of schools in the area with fermented milk, or amasi, was last week ordered to close.

161102dairy01maInnesfree Agriculture and Dairy was identified as a supplier of amasi to a number of schools in the area where over 1000 pupils from 12 schools fell sick after eating amasi and pap.

The dairy was told to shut after Buffalo City Metro’s health services found it to be in contravention of regulations relating to the sale of milk and dairy products which state that the selling of raw milk and raw sour milk for human consumption is prohibited.

In the compliance notice, which is in the Daily Dispatch’s possession, Innesfree was ordered to stop selling dairy products with immediate effect and dispose of any raw milk or raw sour milk.

Innesfree owner Sherene Fourie accused the provincial government of being on a “witch-hunt” and said she was being targeted unfairly.

Fourie said “most” of the schools where pupils fell ill were not on the list of schools supplied by her dairy.

Fourie said health inspectors were wrong to point fingers at her because schools and individual buyers fetched milk from her shop using their own containers. “Did they go to the schools to check if the containers were sterilised? Or whether the food they cook at school was prepared in a proper manner?”

Fourie said some schools bought milk from her and stored it in their own tanks and nobody knew if those tanks were sterilised.

There were hundreds of dairies in the province selling raw milk, she added. “I would like to know, am I the only one who is shut down or are they planning to shut down other businesses?”

Last week 1056 pupils from 11 schools were treated for diarrhoea in three provincial hospitals after eating sour milk and pap offered by the national school nutrition programme.

20 children sickened: E. coli O26 in Italy, 2013

In summer 2013, an excess of pediatric cases of haemolytic uraemic syndrome (HUS) in a southern region of Italy prompted the investigation of a community-wide outbreak of Shiga toxin 2-producing Escherichia coli (STEC) O26:H11 infections. Case finding was based on testing patients with HUS or bloody diarrhoea for STEC infection by microbiological and serological methods.

bobby-crosier-e-coli_-sep_-15A case–control study was conducted to identify the source of the outbreak. STEC O26 infection was identified in 20 children (median age 17 months) with HUS, two of whom reported severe neurological sequelae. No cases in adults were detected. Molecular typing showed that two distinct STEC O26:H11 strains were involved. The case–control study showed an association between STEC O26 infection and consumption of dairy products from two local plants, but not with specific ready-to-eat products. E.coli O26:H11 strains lacking the stx genes were isolated from bulk milk and curd samples, but their PFGE profiles did not match those of the outbreak isolates.

This outbreak supports the view that infections with Stx2-producing E. coli O26 in children have a high probability of progressing to HUS and represent an emerging public health problem in Europe.

Community-wide outbreak of hemolytic uraemic syndrome associated with Shiga toxin 2-producing Escherichia coli O26:H11 in southern Italy, summer 2013

Eurosurveillance, Volume 21, Issue 38, 22 September 2016

C Germinario, A Caprioli, M Giordano, et al.

Dairy cattle feed source of Salmonella

In late 2011 the New Zealand Ministry for Primary Industries reported an increase in confirmed laboratory diagnoses of salmonellosis in dairy herds.

dairy.cull.cattleTo identify risk factors for herd-level outbreaks of salmonellosis we conducted a case-control study of New Zealand dairy herds in 2011–2012. In a multivariable analysis, use of continuous feed troughs [adjusted odds ratio (aOR) 6·2, 95% confidence interval (CI) 2·0–20], use of pelletized magnesium supplements (aOR 10, 95% CI 3·3–33) and use of palm kernel meal as a supplementary feed (aOR 8·7, 95% CI 2·5–30) were positively associated with a herd-level outbreak of salmonellosis between 1 July 2011 and 31 January 2012.

We conclude that supplementary feeds used on dairy farms (regardless of type) need to be stored and handled appropriately to reduce the likelihood of bacterial contamination, particularly from birds and rodents. Magnesium supplementation in the pelletized form played a role in triggering outbreaks of acute salmonellosis in New Zealand dairy herds in 2011–2012.

A case-control study to identify risk factors for acute salmonellosis in New Zealand dairy herds, 2011–2012

Epidemiology and Infection, Volume 144, Issue 10, July 2016, pp. 2154-2164, DOI:

M.A. Stevenson, P.L. Morgan, J. Sanhueza, G.E. Oakley, R.S. Bateman, A. McFadden, N. MacPherson, K.L. Owen, L. Burton, S. Walsh, J. Weston, and R. Marchant


Keep food out of animal education events

Next week I’m tagging along on a field trip with Jack’s first grade class. They’ve been studying the solar system and we’re headed to the planetarium to view the stars and learn about space missions.

No animal exhibits involved in this trip, but I’m sure those are in the future.

I plan on chaperoning any school trips the boys take to the farm, the fair or the petting zoo to help with the onsite risk management.070414.T.FF_.AGEDCENTER1

But, as today’s MMWR highlights, a lot of the disease risk stuff needs to be taken care of before with good planning and procedures.

Yeah, hand washing matters, but so does not letting kids bring lunch/snacks into a contaminated environment.

Or serving food directly in the barn to a 1,000 kids.

Or as Curran et al. say,  ‘These environments should be considered contaminated and should not be located in areas where food and beverages are served’
During April 20–June 1, 2015, 60 cases (25 confirmed and 35 probable) were identified (Figure). Eleven (18%) patients were hospitalized, and six (10%) developed hemolytic uremic syndrome. No deaths occurred. Forty primary cases were identified in 35 first-graders, three high school students, one parent, and one teacher who attended the event. Twenty secondary cases were identified in 14 siblings, four caretakers, and two cousins of attendees.

Food was served inside the barn to adolescents who set up and broke down the event on April 20 and April 24. During April 21–23 approximately 1,000 first-grade students attended the event, which included various activities related to farming. Crude attack rates were higher among those who assisted with setup on April 20 or breakdown on April 24 (three of 14 high school students; 21%) and among attendees on April 21 (22 of 254 students; 9%), than among attendees on April 22 (six of 377 students; 2%) and April 23 (seven of 436 students; 2%).

Animals, including cattle, had been exhibited in the barn during previous events. Before the dairy education event, tractors, scrapers, and leaf blowers were used to move manure to a bunker at the north end of the barn. Environmental samples collected in this area yielded E. coli O157:H7 PFGE patterns indistinguishable from the outbreak strains.

Although it might not be possible to completely disinfect barns and areas where animals have been kept, standard procedures for cleaning, disinfection, and facility design should be adopted to minimize the risk for exposure to pathogens (1). These environments should be considered contaminated and should not be located in areas where food and beverages are served. Hands should always be washed with soap and clean running water, and dried with clean towels immediately upon exiting areas containing animals or where animals have been kept previously, after removing soiled clothing or shoes, and before eating or drinking. Event organizers can refer to published recommendations for preventing disease associated with animals in public settings.

Here’s a set of guidelines we came up with for folks to use when choosing whether to take a trip to these animal events.




Letter: Bacteria in dairy products in baggage of incoming travelers, Brazil

To the Editor: International air travel can lead to the rapid global dissemination of infectious agents. Unlike products and byproducts of animal origin imported between countries under agreements that legally establish sanitary standards, products introduced into a country illegally or irregularly do not follow specific standards and can come from any source, thereby posing a risk to the health status of a country. Animal products transported clandestinely in baggage can contain infectious agents harmful to animal and human health (14). We investigated Brucella spp., Mycobacterium bovis, and Mycobacterium avium subsp. paratuberculosis (MAP) in dairy products seized from baggage of passengers on flights at the 2 main international airports (Guarulhos Airport, São Paulo, and Galeão Airport, Rio de Janeiro, in Brazil.

maxresdefaultDuring 2010–2011, 12 missions were instigated by the International Agriculture Surveillance (VIGIAGRO/MAPA) in airports to detect and seize unauthorized dairy products carried by passengers; 195 products were collected from multiple flights from different destinations. Baggage was scanned by using an x-ray machine and, on detection of a product, was opened by the owner in the presence of a federal agriculture inspector. To avoid contamination, the products were not opened and were sent to the designated Ministry of Agriculture, Livestock and Food Supply Laboratory in their original packaging. All seized products were packed according to the International Air Transport Association standards (5) and transported by commercial aviation with official monitoring to the laboratory.

After completing real-time quantitative PCR (Promega, Madison, WI, USA) using TaqMan technology (Life Technologies, Darmstadt, Germany), we extracted DNA directly from the sample (6,7). The technique for the detection of MAP and eryD Brucella (except strain 19 Brucella abortus) and also using the region RD4 to detect M. bovis were proposed by Irange et al. (8). To detect M. bovis, we used the primers M. bovis-88-F (5′-CGC.CTT.CCT.AAC.CAG.AAT.TG-3′), M. bovis-88-R (5′-GGA.GAG.CGC.CGT.TGT.AGG-3′) and to detect Brucella, we used Bru-Eri-Taq-92-F (5′-GCC.ACA.CTT.TCT.GCA.ATC.TG-3′) and Bru-Eri-Taq-92-F (5′-GCG.GTG.GAT.AAT.GAA.ATC.TGC-3′).

We analyzed 35 containers of dulce de leche, a caramelized milk paste confection, from Argentina (n = 30), Angola (n = 1), and Uruguay (n = 4). We tested all specimens for Brucella spp. and MAP, and 32 for M. bovis. We detected MAP in 1 specimen from Argentina and 1 from Uruguay, Brucella spp. in 3 specimens from Argentina and 1 from Uruguay, and M. bovis in 1 specimen from Argentina.

de-niroThree containers of liquid milk from the United States were collected and analyzed for the presence of MAP; 2 were analyzed for M. bovis and Brucella. Brucella was detected in 1 specimen. Five containers of powdered milk were seized: 2 from Chile, 2 from Angola, and 1 from Portugal. Brucella was detected in 1 container from Chile; Brucella and M. bovis were found in 1 container from Angola. Four containers of yogurt were seized, 1 each from the United States, China, Angola, and South Africa. MAP was detected in those from Angola and South Africa, and the yogurt from South Africa also showed Brucella.

We analyzed samples from 147 cheeses that were confiscated from baggage owned by travelers from 21 countries, mainly from Italy (24.5%), Portugal (22.4%), and France (14.3%). M. bovis was identified in 18 (17.5%) cheeses collected from Italy, Portugal, Spain, the United States, the Netherlands, Lebanon, Morocco, and Norway. MAP was amplified in specimens from 13 cheeses from Spain, United States, Iraq, Israel, Norway, Peru, France, and Portugal, the last 2 countries showed the highest occurrence. Brucella was detected in 62 of the cheeses analyzed, which originated in Bolivia, Chile, Iraq, Lebanon, and Morocco (n = 1 in each country), Netherlands, Israel, and Norway (n = 2 in each country), Turkey and Spain (n = 3 in each country), United States, France and England (n = 4 in each country), Portugal (n=10), and Italy (n = 23).

Both M. bovis and Brucella were detected in 13 (8.8%) cheeses (1 each from Spain, Netherlands, Morocco, and Norway; 4 from Portugal, and 5 from Italy); Brucella and MAP were detected in 4 (2.7%) cheeses (Spain, France, Portugal, and Iraq). Co-amplification of the 3 genes (Brucella + MAP + M. bovis) occurred in 3 (2%) cheeses (United States, Norway, and Portugal). Among the cheeses analyzed, 84 (57.1%) contained isolates that amplified >1 of the genes for the 3 bacteria examined.

Of the 166 dairy products analyzed, Brucella was detected in 70 (42.1%). Cheeses were the most seized products (n = 121) and had the highest number of Brucella-positive results (62/121[51.2%]). Brucella was detected in dairy products that originated in Argentina, Spain, France, Iraq, Israel, Italy, Lebanon, Portugal, and Turkey; it was detected in 4 (21%) of the 19 cheeses from France and in 3 of the 4 (75%) cheeses that originated in Spain. M. bovis was detected in dulce de leche from Argentina, powdered milk from Chile, and in cheeses from Spain, Netherlands, Italy, Lebanon, Morocco, Norway, and Portugal.

Bacteria can be introduced into a country through contaminated animal products that are brought across borders illegally. The risk may be even greater when these products are carried in passengers’ baggage on international flights because of the growing number of international travelers and the wide range of origins of these passengers. Greater attention should be given to agricultural surveillance at airports to mitigate the risk for introduction of these products.

de Melo CB, de Sá MEP, Souza AR, de Oliveira AM, Mota PMPC, Campani, PR, et al. Bacteria in dairy products in baggage of incoming travelers, Brazil [letter]. Emerg Infect Dis. 2014 Nov [date cited].

Author affiliations: University of Brasília, Brasília, Brazil (C.B. de Melo, A.R. Souza, C. McManus, L. Seixas Author affliliations:); Ministry of Agriculture, Livestock and Food Supply (MAPA), Brasília, Brazil (M.E.P. de Sa); MAPA, Galeão Airport, Rio de Janeiro, Brazil (P.R Campani); MAPA, Guarulhos Airport, São Paulo, Brazil (J.O. Luna); MAPA, Confins International Airport, Belo Horizonte/Confins, Brazil (S. Cabral Pinto); MAPA, Brasilia International Airport (BSB), Brasília, Brazil (F.F. Schwingel); MAPA, Pedro Leopoldo, Brazil (A.M. de Oliveira, P.M.P.C. Mota).


Schneider H. Good governance of national Veterinary Services. Rev. Sci. Tech. 2011;30:325–38 [cited 2011 Apr]. PubMed

Hartnett E, Adkin A, Seaman M, Cooper J, Watson E, Coburn H, A quantitative assessment of the risks from illegally imported meat contaminated with foot and mouth disease virus to Great Britain. Risk Anal. 2007;27:187–202 . DOIPubMed

Brückner GK. Ensuring safe international trade: how are the roles and responsibilities evolving and what will the situation be in ten years’ time? Rev. Sci. Tech. 2011;30:317–24 [cited 2011 Apr].

de Melo CB, de Sa MEP, Alves FF, McManus C, Aragão LF, Belo BB, Profile of international air passengers intercepted with illegal animal products in baggage at Guarulhos and Galeão airports in Brazil. SpringerPlus 2014: 3:69.

International Air Transport Association. 3.6.2 Division 6.2—Infectious substances. 2011 January 1 [cited 2011 Aug 10].

Millar BC, Jiru X, Moore JE, Earle JAP. A simple and sensitive method to extract bacterial, yeast and fungal DNA from blood culture material. J Microbiol Methods. 2000;42:139–47. DOIPubMed

Dias NL. Staphylococcus aureus identification, evaluation of the enterotoxigenic potential and methicillin resistance by the PCR technique in dulce de leche samples in the Sete-Lagoas microregion, in the State of Minas Gerais, Brazil (dissertation) [in Portuguese]. Belo Horizonte, Minas Gerais, Brazil: Federal University of Minas Gerais, 2010.

Irenge LM, Walravens K, Govaerts M, Godfroid J, Rosseels V, Huygen K, Development and validation of a triplex real-time PCR for rapid detection and specific identification of M. avium sub sp. paratuberculosis in faecal samples. Vet Microbiol. 2009;136:166–72. DOIPubMed

Emerging Infectious Diseases, Volume 20, Number 11—November 2014

Luna JO, Pinto SC, Schwingel FF, McManus C, Seixas L

Fonterra boosting food safety after last year’s recall

Fonterra Co-operative Group, the world’s biggest dairy exporter, is on track in lifting the quality of its food safety processes, nine months after an independent review into its handling of last year’s false alarm food scare.

UnknownThe Auckland-based company has completed audits of 75 per cent of its plants globally and has embarked on necessary improvements and maintenance where needed, put in place protocols to engage external scientific and diagnostic resources and written food and safety quality into all senior management employment contracts, it said in a statement. It’s also set up an incident management team, created a food safety and quality council, and appointed Greg McCullough as head of food safety and quality.

Careful with that cow inmate: E. coli O111 infections associated with a correctional facility dairy — Colorado, 2010

 Excerpts from an article in today’s U.S. Centers for Disease Control Morbidity and Mortality Weekly Report.

On April 20, 2010, the Colorado Department of Public Health and Environment (CDPHE) was notified by correctional authorities regarding three inmates with bloody diarrhea at a minimum-security correctional facility. The facility, which houses approximately 500 inmates, is a designated work center where inmates are employed or receive vocational training. Approximately 70 inmates work at an onsite dairy, which provides milk to all state-run correctional facilities in Colorado. CDPHE immediately began an investigation and was later assisted by the High Plains Intermountain Center for Agricultural Health and Safety at Colorado State University and by CDC. This report describes the results of the investigation, which determined that the illnesses were caused by Shiga toxin–producing Escherichia coli O111 (STEC O111) infections.

During April–July, 2010, 10 inmates at the facility received a diagnosis of laboratory-confirmed STEC O111 infection, and a retrospective prevalence study of 100 inmates found that, during March–April, 14 other inmates had experienced diarrheal illness suspected of being STEC O111 infection. Pulsed-field gel electrophoresis (PFGE) testing indicated that STEC O111 isolates from inmates matched STEC O111 isolates from cattle at the onsite dairy. An environmental investigation determined that inmates employed at the dairy might have acquired STEC O111 infection on the job or transported contaminated clothing or other items into the main correctional facility and kitchen, thereby exposing other inmates. To prevent similar outbreaks in correctional facilities, authorities should consult with public health officials to design and implement effective infection control measures.

CDPHE staff also inspected the correctional facility’s kitchens and living areas and identified the following conditions conducive to STEC O111 transmission: poor adherence to standard food-service protocols and hygiene practices, including food handlers working while ill with diarrhea; inconsistent availability of hand soap throughout the facility; dairy employees wearing soiled work clothes into the kitchen and living areas; and transport of potentially fecally contaminated lunch coolers and water containers from the dairy into the kitchen.

CDPHE hypothesized that the outbreak was associated with environmental contamination and propagated by person-to-person transmission, possibly through food preparation. On learning of these results, the correctional facility immediately implemented the following public health recommendations: 1) prohibiting potentially contaminated material (e.g., lunch coolers, water containers, and work clothing from the dairy) in the kitchen area, 2) excluding from work all food handlers reporting diarrheal illness since April 1, 3) requiring food handlers with a confirmed STEC O111 test result to have two consecutive negative stool specimens before returning to work, and 4) limiting transfers of inmates to other facilities until they were cleared by the medical staff.

The complete report is available at:

UK farmer fears roosting starlings may cause salmonella in dairy calves, milk

The Brits love their birds.

But not so for a dairy farmer from the Somerset Levels who told BBC News
that roosting starlings and their salmonella-laden poop contaminating feed has led to the loss of 40 calves and is costing his business up to £40,000 a year.

He fears the droppings may also result in salmonella in his cattle’s dairy milk.

Thousands of starlings migrate from Baltic countries, such as Russia, to Somerset and other parts of the UK over the winter months.

In recent years their murmurations as they prepare to roost have become a major attraction for wildlife enthusiasts.

RSPB spokesman, Graham Madge, said, "The fact that starlings are visiting Somerset are not because the RSPB are encouraging them, it’s basically because these birds can find plenty of food in areas that are relatively warm for the winter.”